A perception-based study of the indoor and outdoor acoustic environments in India during the COVID-19 pandemic^a)

The dynamic nature of the COVID-19 pandemic forced governments worldwide into continuously adopting restrictive measures of varying stringency on mobility¹ and introducing public health policies or advisories^2,3 that significantly impacted the lives of individuals and communities and also had a significant effect on the environment, such as reduced air pollution^4,5 and alteration of urban soundscapes^6,7 in a broader sense. Of particular interest to the acoustic community was the sudden and dramatic reduction in outdoor noise levels observed at the onset of the lockdowns in different cities across the globe, which several researchers have attempted to quantify by analysis of data measured on a network of environmental noise monitoring sensors.^8–13 While the environmental noise data are undoubtedly helpful in assessing the effect of prevailing outdoor soundscape on psychological and mental well-being, a direct and, perhaps, readily implementable approach, especially during the challenging pandemic times when outdoor movement was restricted, is to conduct online surveys seeking public opinion on the perceived changes in local indoor and outdoor soundscapes, whether people liked the modified sound environment at their residences, and notably the impact on their day-to-day activities and life.^14–17 Indeed, from a broader perspective, surveys have emerged as a valuable tool for conducting remote acoustic testing.¹⁸ 

Lockdowns forced people to be confined to their homes and to spend more time with family members or flatmates, which may have altered their daily routines and preferences. While the modified outdoor soundscape was generally quieter during lockdowns, it did not necessarily translate into a more peaceful indoor environment, which may be attributed to the particular lifestyle of individuals. For instance, in an otherwise noisy locality that was much quieter during the lockdown, a family may still have experienced increased indoor noise levels due to several factors, such as increased use of home-entertainment, interaction with other members, children playing, etc. Similarly, a person working remotely at a residence in a quiet neighborhood during the lockdown may have experienced increased annoyance and reduced concentration during work due to an unusual increase in indoor noise. In contrast, elderly people who were dependent on aged-care professional help for carrying out their daily activities were forced to live in isolation during the lockdown; the unusually quiet indoor and outdoor environments may not be desirable for them because, unfortunately, this change can increase feelings of loneliness, confusion, and depression among this section of the society. Caniato et al.¹⁴ note that while noise sensitivity is linked with anxiety and depression, low noise levels can result in disagreements and dislikes. Understandably, people exposed to unusually quiet soundscapes for the first time may not perceive it as a pleasant change because it is an unnatural environment for them that is linked to a severe health concern. However, others may welcome the new quieter outdoor and indoor environment irrespective of their home typology or location within the city, regardless of the fact that a life-threatening virus is responsible for the modified soundscape.

In light of the above discussion, it is easy to appreciate a significant research interest in carrying out a purely perception-based soundscape assessment study by means of conducting online socio-acoustic surveys limited to a city^19–23 or conducted nationwide^{15–17,24–26} as well as internationally.¹⁴ An essential objective of these surveys is to quickly assess people's perception of the modified local soundscape caused due to the COVID-19 restrictions while also attempting to understand its effects on their well-being, annoyance, and productivity, among other aspects. For example, a study involving a small group of 109 participants in Antwerp, Belgium, found that people more concerned with the pandemic readily perceived changes in the soundscape of urban green public spaces, such as appreciating the natural sounds and simultaneously becoming more affected by nearby traffic noise.²² Maggi et al.¹⁵ carried out an online survey involving 1371 participants across Argentina and reported that people felt a significant reduction in outdoor noise levels during the lockdown due to a decrease in anthropogenic activities (mechanical sounds) while simultaneously increasing their feelings of tranquillity and happiness. In Italy, an online survey conducted during different stages of COVID-19 lockdowns reported an overall reduction in noise annoyance due to limited traffic and human activities, thereby improving soundscapes.¹⁷ Yildirim and Arefi report that the complaints due to neighborhood noise were reduced by 14% during the COVID-19 pandemic in and around downtown Dallas.²³ In contrast, other survey-based studies highlighted the adverse outcomes of people's perception of the modified acoustic environment in London; they report an increased noise-induced psychological stress or increased rates of complaints due to neighborhood noise.^20,21 A national survey attracting 1053 respondents across Turkey found that the changed local soundscape during the forced lockdown increased annoyance due to concerns of being overheard by neighbors and by noise from one's dwelling.¹⁶ The responses to a survey conducted among people living in multi-unit residential buildings across Canada also suggested a similar outcome.²⁴ 

Based on the responses to an online survey involving 1934 participants across Italy, Puglisi et al.²⁵ reported that 25% of participants who adopted the work from home (WFH) model during the pandemic complained that the noise generated by family members was the primary disturbance source that increased annoyance and loss of concentration, while noise due to neighborhood activities distracted them from completing an assigned task. Torresin et al.¹⁹ conducted an online survey administered to 464 home-workers during the January 2021 lockdown in London, wherein the effect of perception of the indoor acoustic environment on occupants' activity and well-being was investigated. The perception parameters were defined as comfort, i.e., less or more annoyance, and content, i.e., a saturation of environment with sound, while the activities included relaxation and WFH. The response shows that the home environments were perceived as more comfortable and slightly more full of content when they were rated keeping in mind relaxation compared to WFH. At the same time, spaces appropriate for WFH were linked with high-comfort and low-content scores. Furthermore, well-being was linked with more comfortable soundscapes for WFH and relaxation. Therefore, the study recommended that in the future, careful consideration should be given to residential soundscape design aspects that enhance acoustic comfort. In a sequel paper²⁷ based on the same response-set, the authors used a structural equation model to study the influence of acoustical, building, urban, and person-related factors on soundscape dimensions and well-being. The results highlighted the importance of considering the impact of sound typology, building size, availability of a quiet place, number of people at home, and noise sensitivity on the building occupants depending upon the specific work profile.

In India, very stringent lockdown measures were implemented from 24 March up to 3 May 2020 to control the spread of the COVID-19 virus, wherein there was a complete ban on outdoor movement; near-absence of public and private vehicles on otherwise busy roads and highways; a complete halt of commercial operations and construction activities; and closure of schools, colleges, and entertainment places. Slightly less stringent lockdown conditions continued to be enforced from 3 May up to 31 May 2020, while June and July 2020 were the early phases of unlocks (ULs). There were fewer daily cases during the March–May 2020 lockdown and early ULs, while the infection peaked during the first COVID-19 wave in September–October 2020, as shown in the daily cases graph in SuppPub1.jpg.²⁸ The second wave that occurred in April and May 2021 was caused by the Delta variant of the COVID-19 virus producing devastating consequences, with close to 400 000 daily cases. However, the nationwide lockdown measures during these months were far less stringent, possibly due to a reduced fear-psyche among people and also because state governments allowed restricted outdoor movements to meet hospitalization emergencies and medical requirements as well as restricted opening of shops that catered to daily needs. Probably, the experiences gained during the 2020 lockdown made the state governments appreciate that again imposing a complete ban on mobility was neither practical nor sustainable. The third wave that occurred in January and February 2022 due to the Omicron variant resulted in over 300 000 daily cases. However, its effect was very mild, possibly due to vaccination, with very few deaths. For all practical purposes, there were no lockdown measures enforced. A detailed description of the public policies, guidelines, and advisories developed since March 2020 to manage the pandemic can be found on the Ministry of Home Affairs (MHA) website.²⁹ 

The evolving nature of the COVID-19 pandemic and the challenges it posed present a unique opportunity to investigate the changes to the indoor and outdoor soundscapes in Indian cities and towns as well as its effects on people's lives, mainly because the soundscape and socio-economic conditions are vastly different from those of American, European, and UK cities. While such investigations have been reported for different countries,^{14–17,19,21–26} the study period was restricted to the 2020 lockdown duration. In the Indian context, such a study has received only limited attention, where the focus was on changes in the outdoor levels measured^13,30,31 as well as perceived¹³ during the 2020 lockdown and early UL phases. Therefore, the objective of the present work is to assess the perception of the Indian public of changes in indoor and outdoor soundscapes by carrying out a nationwide survey that takes into account their observations during the 2 years from the beginning of the pandemic in March 2020 up to April 2022 (launch of survey campaign) and analyzing the effect on well-being and productivity aspects during WFH and online education.

The paper is organized as follows. Section II describes the questionnaire design and conduct of the online survey and discusses the demographic data of the participants. Section III presents the survey responses, essential conclusions that can be drawn from them, and statistical analyses to support the claims. Section IV presents a discussion where the responses are correlated with anxiety and noise sensitivity, the effect of demographic factors is analyzed, and some salient results are compared with existing studies. Section V summarizes the take-home messages conveyed by this study and some ideas on which future surveys can focus.

An online questionnaire was carefully designed using a Google form in two languages widely spoken in India: English and Hindi (the national language). The survey link was sent to potential participants across the country through emails and social media services, such as Facebook and Instagram, and WhatsApp messenger. The cross-sectional survey comprised a range of questions intended to probe the overall indoor and outdoor soundscape conditions in Indian households during the last 2 years and their effects on mental well-being, productivity for WFH professionals, and online education of school-going children. Furthermore, questions were also asked to assess the severity of anxiety and sensitivity toward noise during the pandemic. In all, there were 20 survey questions broadly arranged into five sections listed in Table I, which also mentions the associated response ratings on a Likert scale. Additionally, the participants were also requested to answer a few personal questions, such as their gender, age-group, highest educational qualification, language in which they usually converse while at home and work, and the city and locality of their residences where they lived most of the time during the last 2 years of the pandemic. Although the survey covered a wide range of topics, it was brief enough to encourage participation. Indeed, most respondents completed the survey questions within 8–10 min. Note that the participation was purely voluntary, i.e., no monetary compensation or in-kind incentives were involved. The survey link was shared from 2 April 2022, and responses were accepted until 2 June 2022, i.e., for 2 months, wherein the participants were requested to carefully take into account the evolving noise scenario and its effects over the last 2 years since the March–May 2020 lockdown. The response data collected on the Google Drive link were downloaded, and the results were processed in Microsoft Excel^® software to produce pie charts or bar graphs based on the question. Excel was also used to conduct a correlation analysis and test whether the responses were statistically significant, which was also confirmed from calculations performed on SPSS^® software, version 26.0 for Windows.

The survey attracted responses from 942 participants across India; SuppPub2.jpg²⁸ shows the survey sample distribution on the map of India, wherein it is observed that all major cities and small towns were covered. However, there was some bias toward Kanpur and Kolkata because these cities had the maximum number of respondents. Statistically speaking, for a 95% confidence level, a sample size n = 942 corresponds to a confidence interval equal to ±3.19%. Note that only healthy adults, i.e., people aged 18 and above with no prior hearing impairments, were requested to respond to the survey link.

Table II shows some important demographic details of the survey responses. An unintentional gender bias was observed; 77% of participants were male, while 21% were female. Furthermore, maximum participation equal to 67.1% was seen from people within the age-group 21–30 who were students attending colleges/universities or young working professionals, while 72.3% of participants had at least a graduation degree. At the workplace, 38.2% and 51.9% of participants conversed in English and Hindi, respectively; however, at home, only 3.6% reported conversing in English and 67.4% in Hindi. In this survey, only 16.8% of participants lived in a metropolitan or tier I city, which includes New Delhi (national capital), Kolkata, Bengaluru, Hyderabad, Chennai, Mumbai, Pune, and Ahmedabad, while the remaining participants lived in smaller cities and towns, i.e., tier II and III categories. Finally, only 16.3% of participants lived in usually noisier commercial areas, while 56.9% of people lived in generally quieter residential areas.

This section aims to understand the prevailing indoor acoustic soundscape at residences during the COVID-19 lockdowns and ULs and its associated mental effects on the participants and their family members by analysis of the responses collectively shown in Fig. 1. It was observed that roughly 77.7% of the surveyed population lived in nuclear families comprising five or fewer members, while the remaining 22.3% lived in a joint Indian family system comprising more than five members as indicated by the response to question 1.1. Generally, the indoor environment in a joint Indian family, especially during lockdown, is expected to be noisy because everyone is at home. However, only 26.2% of participants living in a joint family reported increased noise in their homes. The response to question 1.2 shows that during the last 2 years, only 22.9% of participants felt that the indoor environment was generally noisier than the pre-COVID times, 38.1% of participants did not perceive any significant changes, and 39% thought that the indoor soundscape was indeed quieter. Note that a limitation of the present survey is that it was conducted nearly 2 years after the March–May 2020 nationwide lockdown; despite the unprecedented nature of the event, it is quite possible that people may not have been able to perfectly recall their perception of the local noise environment during the 2020 lockdown as well as the initial phase of ULs. Therefore, the response to question 1.2 may be expected to deliver only an approximate picture of the overall perception of the indoor noise environment during the lockdowns and ULs. To remove this uncertainty and focus only on the noise perception during the 2020 lockdown period, we briefly discuss the response to a question (referred to as Q1.3 in Fig. 1) that was part of the survey conducted during the first half of June 2020 involving 1068 participants across India.¹³ The question asked the participants to rate their local indoor environment during the 2020 lockdown as much quieter, fairly quieter, fairly noisier, and much noisier, which corresponds to a Likert scale from 1 to 4, respectively. The responses to question 1.3 revealed that in contrast to the present (2022) survey, almost 42% of participants felt that the indoor environment was noisier in comparison to the pre-lockdown times, while the remaining participants felt that it was quieter in general. Now, the mean score $μ$ and standard deviation $σ$ of the responses for question 1.3 were 2.33 and 0.81, respectively, while the counterpart values for question 1.2 (recalibrated on a 1–4 scale) were equal to 2.21 and 0.84, respectively. (Henceforth, the standard deviation will not be reported for brevity.) In this work, the significance level is set to probability p = 0.05, i.e., a two-tailed p-value less than 0.05 will signify a statistically significant result. In other words, the null hypothesis, which assumes that two sets of variables are uncorrelated, is rejected in favor of the alternate hypothesis, which agrees that there exists a correlation or link between the variables. By carrying out a two-sample unequal variance t-test between the responses to questions 1.2 and 1.3, it was found that the difference between the mean scores is large enough to be statistically significant, probability p < 0.001. (The t-value or test statistic will not be reported in this work.) This seems to suggest that a slight difference may indeed exist between the mean scores of the perceived indoor noise levels for the two surveys, and the difference in the time frames may have caused this peculiarity.

The response to question 1.2 (present survey) may be explained by noting the bar graph responses to question 1.4, which suggest that 48.7% of participants felt that the noise due to entertainment sources increased, which is understandable because people were confined to their homes, especially during the 2020 lockdown, which resulted in increased viewership of television/movies, use of music systems, playing video games, etc. Simultaneously, 59.1% of participants observed an increase in indoor noise contributions from video-calling with relatives/friends by their family members, while 62.4% of participants felt an increase in indoor noise due to greater family interaction. The above observations are in general agreement with the responses that nearly 44% and 46.4% of participants felt a general decrease in neighbor noise level and activity of children, respectively. Furthermore, 52.7% of participants did not perceive any changes in the indoor noise levels due to home appliances, while almost 50% of people appreciated an increase in the natural sounds at their residences due to fauna and other biological sounds.

Questions 1.5 and 1.6 investigate the psychological or mental well-being aspects of changes in the indoor soundscape. The responses to the former question were subdivided into two categories, namely, (1) 216 participants who perceived an increase in indoor noise levels during the COVID times and (2) 726 participants who either did not observe any change or felt that the levels decreased, as shown in pie charts presented in Fig. 1 [Q1.5 (a) and (b), respectively]. Within the first category, it was observed that nearly 51% of participants felt claustrophobic, while in the second category, only 34.7% of participants experienced claustrophobia. The mean scores $μ$ on the Likert scale were equal to 3.51 and 3.16 for the first and second categories, respectively. By carrying out a t-test between the two categories of respondents, the result was found to be statistically significant (p < 0.001). Therefore, the perception of a noisier indoor environment during COVID-19 times is linked to increased feelings of claustrophobia among the residents, although confinement or home isolation for several days during lockdowns would have somewhat contributed to the same.

On the other hand, the responses to question 1.6 show a mixed response for aspects such as peace and happiness; the proportion of people who perceived an increase or decrease in these aspects was comparable during the last 2 years of the pandemic. The unprecedented situation caused by the 2020 lockdown and the first few UL phases was generally stressful, as many feared getting infected and going into self-isolation. At the same time, for some sections of society, there was considerable stress due to the loss of jobs, i.e., loss of a regular income source and associated uncertainties, which would have affected peace and happiness while adding to feelings of fear, worry, and despair. Additionally, 44.6% of participants reported that their concentration levels decreased due to indoor noise levels, which is consistent with the result that 44.1% experienced a somewhat greater annoyance due to changes in the indoor soundscape.

Figures 2(a)–2(c) show the response to the questions about the outdoor noise environment near residences and its effect on the mental well-being of residents. To begin with, the participants were first requested to compare the outdoor noise levels perceived during the strict lockdown in March–May 2020, the different UL phases in 2020, the less stringent lockdown in April–May 2021, and ULs in 2021 up to April 2022 with the noise levels observed during pre-COVID period (see question 2.1). As noted in Table I, the comparison of their perception was quantified on a Likert scale ranging from 1 to 10, where a rating of 1 denotes much lower, 5 indicates nearly the same level, and 10 indicates much higher as compared to the pre-COVID times. Based on the responses for each period, a boxplot distribution is presented in Fig. 2(a) for all participants, where the overall survey results are shown in blue, and the results only for those participants living in New Delhi National Capital Region (NCR) region are shown in red. For the overall survey, it is readily observed that the perceived outdoor noise levels were significantly lower during the 2020 lockdown. This is consistent with the survey findings reported in Mimani and Singh,¹³ in which one of the questions asked the participants across India to rate their local outdoor environment during the 2020 lockdown as much quieter, fairly quieter, fairly noisier, or much noisier. On a Likert scale, these options ranged from 1 to 4, respectively, and the mean score $μ$ of the responses was 1.63. Now for comparison, the responses to question 2.1 of the present survey on the outdoor noise perception during the 2020 lockdown were recalibrated on a 1–4 scale, whereby the mean was found to be equal to 1.62. Using a t-test, it was found that probability $p = 0.787 ,$ i.e., the difference between the means of the 2020 and 2022 surveys is not large enough to be statistically significant, i.e., the null hypothesis cannot be rejected that states that there is no difference between the mean scores of the outdoor noise perception during the 2020 lockdown given by both surveys. This suggests that even after more than 2 years, respondents still seem to have profound memories of the unusually quiet local outdoor noise environment during the strict lockdown in March–May 2020, which was a completely unprecedented event for most people.

Furthermore, the bar-graph responses to question 2.2 indicate that 73.1% and 77.7% of participants, respectively, felt that public transport and local street noise was reduced significantly during the 2020 lockdown as well as early phases of ULs, while 66.9% and 63.8% of participants, respectively, were of the view that the industrial and construction noise also reduced during the same period. Generally speaking, the responses about outdoor soundscape during the 2020 lockdown agree with the common knowledge that due to severe restrictions on outdoor movement, there was a near-absence of traffic and a complete halt of commercial and industrial operations during the strict lockdown.

During the ULs in 2020, the perception was that the levels gradually increased, which was consistent with the policies of the state governments to only gradually and partially lift movement restrictions during ULs. However, during the April–May 2021 lockdown, people perceived the levels to be comparable with the pre-COVID times, which is quite understandable given its partial or less stringent nature of enforcement. Now, a complete UL period began from June 2021 onward, during which the situation was practically the same as during pre-COVID times; as mentioned earlier, there were no restrictions whatsoever on outdoor movement, including interstate travel, and educational institutions and commercial activities were fully allowed. During this period from June 2021 up to April 2022 (starting month of the survey campaign), people perceived the levels to be slightly more than the pre-COVID times [see Fig. 2(a)]. Note that following the end of the 2021 lockdown, the situation quickly returned to normal; in fact, during the third wave in January 2022, there were no mobility restrictions at all, which is also the case presently. Therefore, during the last several months of the year 2022 as well as in the near future, it is anticipated that people will perceive the local outdoor noise levels to be comparable to the pre-COVID times, provided that there is no further outdoor movement restriction.

The responses to questions 2.1 and 2.2 are further corroborated by Figs. 2(d) and 2(e), which, respectively, show the weekly trend, i.e., the 7-day moving average of the equivalent daytime noise level $L Day d B ( A )$ variation at Anand-Vihar, a commercial location in New Delhi, and the average traffic congestion level in the same city expressed as a percentage for the years 2019, 2020, and 2021. Note that the $L Day d B ( A )$ data are freely available on the Central Pollution Control Board (CPCB) website³² and were obtained from the noise levels recorded by a noise monitoring terminal operating under the National Ambient Noise Monitoring Network (NANMN); see Garg et al.³³ for details. On the other hand, the traffic congestion data were taken from the website of TomTom,³⁴ a multinational developer of maps and global positioning system (GPS) location technology, which presently provides real-time traffic congestion monitoring for four Indian cities, namely, New Delhi, Bengaluru, Pune, and Mumbai, among other cities worldwide. As reported previously,¹³ Fig. 2(d) readily demonstrates that the March–May 2020 lockdown resulted in a significant reduction of up to 12 dB(A) in outdoor noise levels as compared to the ambient levels observed during the same months in 2019. Furthermore, the levels increased almost linearly during 2020 ULs up to the end of September 2020; they were noticeably higher in October 2020 due to the festive season, and from mid-November 2020 up to mid-April 2021, they were comparable with 2019, i.e., the pre-COVID graph. The above observations are well-supported by Fig. 2(e), where one can compare the variation of traffic congestion bar graphs of the year 2020 for the months March–December with the counterpart bar graphs for the year 2019. Figure 2(d) also shows that during the lockdown from mid-April to the end of May 2021, there was only a minor reduction in noise levels compared to the results during the same time in 2019, which is again corroborated by the traffic congestion bar graphs for the year 2021. For the complete UL, which started in June 2021 onward up to the end of April 2022, the levels exhibited daily random fluctuations and were more or less comparable with the 2019 graph except during October 2021, when the levels were marginally higher due to festival season. The video presented in SuppPubmm1.mp4²⁸ shows the daily variation of the weekly trend of $L Day d B ( A )$ at 10 NANMN stations shown on a map of New Delhi, which includes four commercial (C), two residential (R), and four silence (S) zones for the years 2019 (pre-COVID), 2020, 2021, and up to April 2022. With the exception of two silence zones and a commercial location, the same comments as noted for the Anand-Vihar station may also be made for most NANMN locations, where a significant reduction in levels was observed during the 2020 lockdown. However, during the 2021 lockdown period, only a small drop in the levels was observed at these locations, while beginning from the complete UL in June 2022 up to April 2022, the levels exhibit daily random variations comparable with the 2019 pattern. Indeed, the noise variation at these locations compares well with the outdoor noise perception of New Delhi NCR participants shown in Fig. 2(a).

The aforementioned discussion is summarized by noting an interesting observation: maximum reduction in outdoor levels was observed only when the restrictions were most stringent (April 2020), while relatively far less reduction was observed when the less stringent 2021 lockdown was imposed. However, the benefit in terms of noise pollution control was clearly temporary because the levels returned back to the pre-COVID levels either gradually or quickly based on the government policies to lift mobility restrictions and, importantly, people's fear-psyche of being infected with COVID-19. Therefore, in the future, if no restrictions are imposed, the levels will most likely be comparable to those of pre-COVID times.

In contrast to the indoor noise effect on well-being, the response to question 2.3 suggests that due to an overall reduction in outdoor environmental noise during lockdowns, 56.9%, 43.6%, and 45.6% of participants experienced, in general, an increase in peace, happiness, and concentration, respectively, while 34% experienced less annoyance due to outdoor noise sources. Based on the Likert scale defined in Table I, the mean scores $μ$ for peace, happiness, concentration, and annoyance levels due to outdoor noise were found to be equal to 3.53, 3.32, 3.31, and 2.94, respectively, while the counterpart values for the indoor noise case are given by 3.05, 2.94, 2.75, and 3.20, respectively. For a given mental well-being aspect, a paired two-sample t-test revealed probability $p < 0.001 ,$ i.e., the difference between the means is large enough to be statistically significant. Therefore, the present survey suggests that during the lockdowns, the quieter outdoor noise environments in Indian cities were linked to an increase in feelings of peace, happiness, and concentration and simultaneously a marginal reduction in annoyance, while the indoor noise environments were linked to slightly decreased concentration and increased annoyance but had little effect on peace and happiness. Indeed, a relatable result was observed from the survey responses of a previous study;¹³ the respondents were asked to rate on a scale of 1 (no issues) to 5 (great concern) the severity of health-related problems due to outdoor noise before and during the lockdown. The mean score $μ$ before and during the 2020 lockdown was 3.43 and 2.40, respectively, which was statistically significant. Therefore, the 2020 survey also indicated that quieter outdoor environments were linked to improved public health and well-being. Finally, a result worth noting is the sleep pattern due to changes in the indoor and outdoor soundscape; the responses to questions 1.6 and 2.3 indicate that roughly the same percentage (46%) of participants experienced an enhanced sleep pattern, while 31.1% and 40.1%, respectively, were not affected. However, a t-test showed no statistically significant difference between the mean sleep pattern scores due to changes in outdoor and indoor noise sources.

Figure 3 shows the responses to the survey questions, which aim to study the effect of the local indoor soundscape on different productivity aspects during WFH. Based on the bar-graph responses to question 3.1, it is evident that, on average, 57.3% of participants experienced a general reduction in productivity aspects, such as time management, efficiency, self-discipline, and work-life balance. Furthermore, 48.6% of participants reported that their ability to meet the target deadlines was adversely affected. On the other hand, a roughly equal percentage of participants reported either an increase or decrease in cognitive skills. Additionally, the reduction in the aforementioned productivity aspects does not seem to be influenced by educational qualifications based on which a participant may have a certain job profile that can be carried out with ease or difficulty from a home office. Given these responses, it appears that in the Indian context, the overall productivity is reduced during WFH due to indoor noise levels, resulting in increased working hours at home as reported by 61.3% of participants; see the response to question 3.2.

While the increase in indoor noise levels due to entertainment sources, children playing, and family interaction may have contributed to the above experience, it is equally possible that several distractions at home, such as people visiting as well as a semi-formal work environment, may have been equally responsible. Furthermore, the response to question 3.3 suggests that the background noise emanating from the other end during online meetings (Zoom calls, Google-meet, etc.) would have most likely degraded the home environment acoustics, resulting in adverse effects on hearing, listening, and other cognitive abilities, thereby rendering it challenging to work in the home office, as reported by nearly 71% participants. The response to question 3.4 suggests that 47.6% of participants were sometimes distracted, while 25% experienced more frequent distractions due to indoor noise sources while working from home. From an overall point of view, only 15% of participants prefer that the WFH model be implemented in the future, 49% prefer a hybrid mode of work, and 36% prefer working in office as indicated by the response to question 3.5. In the Indian context, these responses are in contrast with the popular trends reported previously, which indicated that several multinational companies were contemplating implementing a WFH model in the long run for their employees as it would help them significantly save on commute time and costs and on office resources and reduce their day-to-day expenses.^35,36 However, it is essential to clarify that workplace preference is not influenced solely by the local noise environment; instead, in general, it largely depends upon the nature of the work and, to some extent, on consideration of one's conveniences and personal circumstances, which allow employees to use a home office on some days while physically reporting to work on the remaining days. For example, a software professional may prefer WFH on most days of the week to save on costs and time while still delivering results and meeting online with supervisors and clients.

The changes to the home soundscape during the lockdowns and ULs, nevertheless, affect the productivity factors (question 3.1) and concentration levels (questions 1.6 and 2.3), which can indeed influence an individual's preference for WFH, work in office, or hybrid mode. Therefore, statistical analysis between the workplace preference and productivity aspects can help establish a link between the two, i.e., determine whether they are related or not. To this end, the mean scores were computed for each productivity aspect and concentration level (see Table III) along with the standard deviation for participant groups preferring WFH, working in office, and hybrid mode. The data were then used to conduct a one-way analysis of variance (ANOVA) test among the three categorical variables for different productivity aspects and concentration. The results were found to be statistically significant (p < 0.001) for all aspects, which was followed by carrying out a post hoc analysis. In this case, the post hoc analysis comprises three independent groups of t-tests with a Bonferroni corrected significance level given by $p = 0.05 / 3 ≈ 0.017$⁠. Again, for a given productivity aspect, the results for individual t-tests were found to be statistically significant, which implied that the alternate hypothesis holds, i.e., the mean scores for different categories are not the same.

From the values reported in Table III, it may then be argued that the changes in indoor soundscape tend to hardly affect the productivity aspects and concentration of respondents who prefer WFH mode, which is quite understandable. On the other hand, these factors tend to noticeably decrease for participants who prefer working in office. However, the mean scores for concentration levels increased due to outdoor noise reduction for all categories, particularly for participants preferring WFH. Regarding working hours at home (question 3.2), the mean scores for participants who prefer WFH, hybrid mode, and working in office were given by 2.5, 2.34, and 2.32, respectively. Similarly, insofar as a distraction due to indoor noise sources is concerned (question 3.4), the mean scores for respondents preferring WFH, hybrid mode, and working in office were given by 1.73, 2.03, and 2.19, respectively. For the responses to these two questions, a one-way ANOVA test followed by a post hoc analysis showed that the results were statistically significant, thereby implying that (1) the respondents who preferred working in office were somewhat more inclined to feel that the daily number of hours put in by them increased due to the existing soundscape at their homes, and (2) those who preferred WFH tended to be somewhat less distracted due to indoor sources.

In light of this discussion, one may conclude the home noise environment indeed influenced workplace preference through its effect on different factors, such as productivity, number of daily working hours, and tendency to feel distracted.

Figure 4 shows the responses to the survey questions, which attempt to understand the effect of the indoor acoustic environment during the last 2 years on aspects of online education of school-going students. Note that this survey section was completed by only 318 participants who had children or observed their behavior at home during the pandemic. Additionally, English was the medium of instruction for 219 children, while Hindi was the medium of education for children of the remaining 99 participants.

The bar-graph responses to question 4.1 indicate that 79.8%, 78.6%, and 77.4% of participants felt that involvement, concentration, and attitude or motivation, respectively, of their children toward studies was noticeably reduced due to the detrimental effect of the prevailing indoor soundscape, while 79.6% of participants attributed it to a reduced one-on-one teacher-student focus. Furthermore, the mean scores for these parameters were given by 2.1, 2.0, 2.0, and 2.0, respectively, indicating a reduction. The above responses are consistent with the outcomes of an Australia-wide survey carried out by University of Melbourne researchers, who report that online education had a rather disengaging effect on vulnerable students, and there were concerns regarding a lack of student-teacher communication and well-being for students facing challenging personal circumstances.³⁷ Similarly, an online survey conducted across American universities suggests that a sudden shift from in-person to online classes during Spring 2020 resulted in increasingly reduced satisfaction and engagement among UG and graduate students as the semester progressed, while the UG students experienced feelings of frustration and reduced engagement and accountability during remote learning.³⁸ 

The responses to question 4.2 indicate that the children of 38.7%, 31.4%, and 58.5% of participants experienced a general increase in anxiety due to uncertainty $( μ = 3.2 )$⁠, depression $( μ = 3.1 )$⁠, or lethargy or boredom $( μ = 3.5 )$⁠, respectively, owing to the nature of online instruction amidst the noisy indoor soundscape. Furthermore, 66.3% of participants reported a reduction in physical activity due to a lack of freedom of movement, and 40.6% reported a simultaneous increase in eye issues faced by children due to long hours of sitting in front of the computer screen. Therefore, based on the observed noise level at homes, it was observed that over 66% of participants were in general disagreement that in the long-term, online education could be a suitable alternative to in-person classes; rather, they preferred that instruction should be given in a regular or physical class (refer to the pie-chart response to question 4.3). The responses to question 4.4 show that 62% did not prefer for their children to engage in a flexible or hybrid-form of education, which is also indicated by the mean score given by 3.70. Indeed, this result is well-anticipated given the responses to questions 4.1 and 4.2. Furthermore, the link between a parent's decision whether to continue the hybrid mode of education or not with different factors mentioned in questions 4.1 and 4.2 is investigated using the Pearson correlation analysis. Concentration, attitude toward studies, and physical activeness were the only factors that were found to be statistically significantly correlated with the parent's decision; the correlation coefficient r for these factors is given by –0.097, –0.132, and –0.089, respectively, thereby indicating a weakly negative correlation. This suggests that a parent's preference for online or physical classes for their children was indeed influenced by aspects of online study and behavior, which, in turn, are affected by the home noise environment changes.

To quantify the effect of anxiety experienced by the residents from the beginning of the 2020 lockdown to the end of the 2021 lockdown, a Generalized Anxiety Disorder (GAD-7) assessment³⁹ was carried out. The GAD-7 assessment is a seven-item instrument that measures the severity of the anxiety disorder; see question 5.1 in Table I. To this end, the participants were asked to answer each question by selecting one of the following responses: not at all, several days, more than half the days, or nearly every day, where these options correspond to a numerical score of 0, 1, 2 and 3, respectively. The severity of the anxiety was represented by the sum of individual scores in the range of 0–4, 5–9, 10–14, and 15–21, signifying minimal, mild, moderate, and severe anxiety, respectively. The responses suggest that 30.3%, 36.1%, 21.4%, and 12.1% of the participants experienced minimal, mild, moderate, and severe anxiety symptoms, respectively.

For each participant, the mental well-being, as well as productivity aspects, were assigned a score on the Likert scale, as indicated in Table I. Following this, a Pearson correlation analysis was carried out to obtain the correlation coefficient $r$ between the anxiety severity and mental well-being aspects for indoor and outdoor soundscapes, as well as between anxiety severity and productivity aspects for the entire sample of participants. Table IV indicates that the mental well-being aspects, such as peace, happiness, concentration, and sleep, as well as all productivity aspects are negatively correlated with anxiety severity as expected, although the correlation is somewhat weak. On the other hand, annoyance and feelings of claustrophobia have a weak positive correlation with anxiety severity. Furthermore, the p values indicate that the correlation between anxiety and different mental aspects, as well as between anxiety and productivity aspects, is statistically significant.

Similarly, residents were also requested to rate their sensitivity toward seven different noise sources mentioned in question 5.2 of Table I under general conditions. For each noise source, the responses were measured on a Likert scale ranging from 1 (very low) to 5 (very high). The total of individual scores was calculated, representing the overall sensitivity score for the concerned participant. (The minimum and maximum sensitivity scores were equal to 7 and 35, respectively.) Pearson correlation analysis was carried out between noise sensitivity and mental well-being for indoor and outdoor soundscapes. It was observed that peace, happiness, concentration, and sleep are negatively correlated with noise sensitivity. As anticipated, the correlation is much weaker for the case of the outdoor soundscape as compared to the indoor one. Furthermore, note that the correlation between concentration and sleep is not statistically significant concerning the outdoor soundscape. Finally, the Pearson correlation analysis was also carried out between anxiety disorder during the pandemic period and noise sensitivity, wherein it was found that the correlation coefficient $r = 0.143$ and p < 0.001, i.e., they are positively correlated as expected, and the result is statistically significant.

In light of this discussion, it may be concluded that those with a greater sensitivity toward noise tend to become slightly anxious when the soundscape changes. Also, the different attributes of mental health and productivity were mildly affected for individuals who experienced anxiety issues when their local soundscape changed during the pandemic.

Based on the demographic information of the participants, some interesting inferential statistics were observed regarding the effect of changes in the local soundscape during the last 2 years of the pandemic.

1. 28.9% of females perceived that the indoor soundscape was noisier during the lockdowns in comparison to only 21.5% of males who felt the same—this is consistent with the finding that nearly 56% of females felt that the home-entertainment sources were noisier in comparison to only 47% males who felt the same. 46.4% of females reported a reduction in peace, happiness, and concentration; in contrast, only 38.2% of males noted a decrease in these mental well-being aspects. Similarly, 52.9%, 42.6%, and 66.7% of females said they experienced an increase in annoyance and claustrophobia and had to put in more working hours, respectively, in comparison to only 41.6%, 37.7%, and 54.8% of males who experienced the same issues, respectively. These statistics suggest that females were slightly more prone to be affected by changes in the local soundscape. The above results are supported by observing the noise sensitivity scores of female and male participants given by 22.0 and 20.8, respectively. The above scores were statistically significant, as confirmed by the t-test (p = 0.003). A chi-squared test was also carried out to investigate whether gender and workplace preference are related; it was found that $χ 2 = 0.54$ and p = 0.76, thereby implying that the null hypothesis cannot be rejected, i.e., workplace preference is independent of gender.

2. For participants belonging to the age-group 21–30, nearly 51% preferred to work in a hybrid mode, 33.5% preferred to return to the conventional style of working from offices, and only 15.5% preferred WFH. In contrast, notwithstanding the much smaller sample of participants in the age-group 41 and above, only 33.3% chose to work in a hybrid mode, 14% preferred WFH, and 52.6% liked working from offices. This is despite the result that 57.1% of participants within the age-group 21–30 reported a reduction in productivity aspects such as time management, efficiency, self-discipline, and work-life balance, while only 41.2% of participants in the age-group 41 and above reported a reduction in these aspects. To investigate whether workplace preference and age-group are related, a chi-squared test was conducted, which showed that the statistic $χ 2 = 9.50$ and p = 0.049, thereby implying that the alternate hypothesis is valid, i.e., workplace preference and age-group are indeed not independent. Therefore, given the above results, it may be stated that the younger workforce preferred to work in a hybrid mode, possibly due to the greater freedom and flexibility offered by this work model, notwithstanding the adverse effects on productivity due to changes in indoor noise environment.

3. 38.9%, 39.7%, and 47.5% of participants belonging to the age-group 21–30 reported a decrease in peace, happiness, and concentration, respectively, due to indoor soundscape changes in contrast to only 12.9%, 17.5%, and 8.8% participants, respectively, who were aged 41 and above. Furthermore, 42.3% of participants within the 21–30 age-group experienced increased annoyance compared to 35.1% of participants aged 41 and above. This suggests that the mental well-being aspects of the younger population are more likely to be affected by a possible increase in indoor noise levels. The result is supported by noise sensitivity scores of these two age-groups, which were found to be statistically significant (p = 0.049); the score was 21.3 for participants aged between 21 and 30 years, while the score was 19.9 for those aged 41 and above.

4. The mean scores of working hours for participants who converse in English and Hindi at the workplace were 3.69 and 3.55, respectively. Similarly, the mean noise sensitivity scores were 21.4 and 20.7 for participants speaking in English and Hindi, respectively. Again, t-tests confirmed that the above scores were statistically significant, based on which one can conclude that the English-speaking workforce was marginally more affected by the indoor soundscape. However, for productivity aspects and mental health parameters, the comparison of mean scores for English- and Hindi-speaking participants was not statistically significant.

5. The percentage of students with English or Hindi as the primary medium of instruction who experienced a reduction in involvement and concentration and increased feelings of depression was roughly the same. This suggests that the medium of teaching did not influence online studies.

This section aims to briefly present a few interesting findings reported in previous articles on people's perception of changes to the local soundscape due to lockdowns and their effect on mental well-being and remote working and compare them with the present study.

Based on the responses to an international survey, Caniato et al.¹⁴ found that 75% of respondents worldwide and over 95% of Italian participants perceived that the outdoor environment was quieter during lockdowns. The prevailing outdoor soundscape influenced the indoor noise, so at least one-third of respondents worldwide and 60% of Italian participants felt that their homes were noisier than usual, presumably because residential buildings were more crowded. In comparison, for the Indian scenario, during the 2020 lockdown, over 90% of respondents felt that the outdoor environment was quieter,¹³ while 42% of participants felt that their homes were noisier. Caniato et al.¹⁴ also note that the age difference played an important role in indoor soundscape preference; younger respondents aged 40 or less preferred noisier places, while older ones aged 60 or above showed less preference for the same. Furthermore, the well-being of women living in crowded homes was more likely to be affected compared to men, who did not express much concern. In the present study, the responses reveal a similar result—the younger workforce preferred a hybrid mode of work despite a noticeable drop in productivity, possibly due to noisier indoor environments at Indian homes, which also affected the mental well-being of females marginally more than the males.

Concerning noisier homes, Lee and Jeong²¹ reported that noise complaint tweets in London during the 2020 lockdown were more than twice as numerous compared to the tweets in 2019, and more specifically, tweets complaining about noise from neighbors during the lockdown were nearly four times more numerous than in 2019. The tweets were also supported by a questionnaire survey answered by 183 participants based in London, which suggested that in comparison to the pre-lockdown period, roughly 10% more respondents during the lockdown felt highly annoyed due to neighbor noise arising from airborne sources such as talking/shouting and watching television or playing loud music. In contrast, the present study shows that only 26.3% of the participants across India perceived a noticeable increase in neighbor noise, possibly because their homes were much louder during the lockdowns.

Dumen and Saher¹⁶ reported that during the 2020 lockdown, the noise sensitivity for Turkish residents was positively correlated with anxiety scores (r = 0.101, p < 0.001) as well as with stress scores (r = 0.092, p < 0.003). They also noted that the noise annoyance score (on a scale of 10) due to traffic decreased from 3.79 to 2.67, but simultaneously, the dwelling noise annoyance score slightly increased from 2.68 to 2.94. In the present study too, during the lockdowns, noise sensitivity and anxiety were positively correlated (r = 0.143, p < 0.001), while the annoyance due to outdoor noise was slightly reduced, but due to the indoor soundscape, it marginally increased as noted in Sec. III B.

Maggi et al.¹⁵ reported that in category III cities of Argentina with more than 1 × 10⁶ inhabitants, 10%, 19%, and 55% of the respondents experienced happiness, tranquillity (peace), and irritation (annoyance), respectively, before the 2020 lockdown. However, during the lockdown, the percentage of respondents who experienced happiness and tranquillity increased to 30% and 83%, respectively, while those who perceived irritation decreased to only 10%. In the present survey, for tier I Indian cities (population greater than 10 × 10⁶), it was found that 58.6%, 48.7%, and 48.0% of respondents experienced an increase in peace, happiness, and concentration, respectively, while 38.8% felt a reduction in annoyance.

Aletta et al.⁴⁰ carried out a traffic simulation and noise emission assessment for Rome to investigate the impact of mobility restrictions on road noise pollution. Their results showed a significant decrease of 64.6% and 34.3% in traffic volumes during the first and second lockdown phases, respectively, which accounted for noise reduction on the road network. TomTom traffic index data showed that during April 2020, the traffic congestion in Indian cities like New Delhi, Pune, Bangalore, and Mumbai was 6% or less, which was an unprecedented reduction. Naturally, over 73% of the respondents in the present survey said that traffic noise pollution was reduced during the 2020 lockdown.

Bartalucci et al.¹⁷ conducted an online survey involving 400 participants across Italy, which, combined with a logistic regression model, showed that the perception of traffic noise had increased during the March 2020 lockdown, especially for those over 35 years of age. In the Indian context, the preset survey responses suggest an opposite trend; the mean score of outdoor noise perception for younger participants within the age-group 21–30 was found to be equal to 4.17, while that for older participants aged 41 and above was given by 3.31, and the results were statistically significant. However, note that the type of area where the participants lived can also significantly influence the noise perception. A one-way ANOVA was applied to the outdoor noise perception responses for the 2020 lockdown for the three categories of participants living in commercial, residential, and rural areas, wherein no statistical difference was found. This seems to suggest that younger respondents tended to have a greater outdoor noise perception during the lockdown, which was independent of the area where they lived.

In a recent article, Kumar et al.⁴¹ analyzed the effect of lockdown during the second wave of COVID-19 on environmental noise by analyzing data measured at 25 real-time ambient noise monitoring stations located across New Delhi that were installed by the Delhi Pollution Control Committee (DPCC). (The exact location and the area where each noise monitoring station is located, along with daily noise metrics, is available on the DPCC website.⁴²) Except for Karol Bagh, all other DPCC noise monitoring sites experienced a minor reduction in noise levels of less than 5 dB(A) during the April–May 2021 lockdown with respect to the pre-lockdown period. Furthermore, except for a couple of industrial zone sites and a silence zone site, it was reported that almost all DPCC sites experienced only a minor increment in noise levels within the range of 0–5 dB(A) during the post-lockdown period, i.e., from June 2021 onward, with respect to the 2021 lockdown. In this work too, it was observed that during the second wave, there was only a small reduction in the levels at several NANMN locations across New Delhi as compared to the pre-lockdown period, while during the ensuing UL period, the levels exhibited daily random fluctuations that were comparable to the pre-lockdown times; see SuppPubmm1.mp4²⁸ and Fig. 2(d). Furthermore, the NANMN data also suggested that the third wave from mid-January 2022 up to the third week of February 2022 did not have any impact whatsoever on the environmental noise in New Delhi. In fact, the above results corroborate people's perception of outdoor noise levels in New Delhi during the second wave and complete UL period shown by the box plot in Fig. 2(d).

In the aforementioned article,⁴¹ the results of a survey from 810 participants in Delhi city were also presented. The response to one of the questions revealed that only 10.8% of the participants felt that the local street noise and car parking noise were the dominant noise sources during the second wave lockdown, while 41.1% and 15.4% perceived that the outdoor noise due to police vans and ambulance sirens and to loudspeaker public address systems, respectively, were the dominant sources in the same period. These numbers not only point toward the partial nature of the 2021 lockdown but also indirectly corroborate the responses to question 2.2, which revealed that 73.1% and 77.7% of participants felt a general reduction in public transport noise and local street noise, respectively. Furthermore, it was reported that due to domestic noise sources (appliances, televisions, music systems, etc.) during the lockdown, 28.6% of participants said that they felt highly annoyed, while 31.5% said that they were moderately annoyed. In comparison, 11.3% and 32.8% of the participants across India reported a marked increase and slight increase, respectively, in annoyance due to the changed indoor soundscape during the 2020 and 2021 lockdowns.

Puglisi et al.²⁵ investigated the effect of home soundscape on work productivity and mental well-being during the lockdown in Italy from March to May 2020 by asking respondents to rate on a scale of 1 (strongly disagree) to 5 (strongly agree) whether the indoor noise (a) caused frequent work interruptions, (b) did not allow them to meet their daily work completion targets, or (c) affected their work performance and well-being. Participants who were not sensitive to home noise environment, regardless of whether they worked in a shared or separate environment, had a mean score of less than 2 for the above questions, which implied that they tend to disagree that the noise could affect their work and well-being or cause annoyance. The participants who were sensitive but worked in a separate workspace were found to have a mean score $μ ≈ 2.25$⁠, implying that they also somewhat disagreed. However, for those participants who used a shared space during WFH, their mean scores were higher (between 2.55 and 2.8), while the difference with respect to the participants who worked in a separate room was statistically significant, thereby indicating that the former category of participants perceived higher noise annoyance and a slight reduction in work productivity and harmony in interpersonal relationships at home. In the present study, participants with a sensitivity score (see Sec. IV A) less than 21 were considered not sensitive to noise, while those with a score of 21 or above were deemed sensitive. Based on their responses to question 5.2, it was found that 363 respondents, i.e., 38.5%, were not sensitive to noise, while the remaining 579, i.e., 61.5%, were noise sensitive. The mean scores of productivity aspects, such as time management, efficiency, self-discipline, meeting deadlines, and work-life balance, were within the range of 2.5–2.65 for both sensitive and non-sensitive participants, and statistically no significant difference was found between the two groups. Therefore, in the Indian context, the responses suggest that work productivity was somewhat affected by the local soundscape changes; however, it was found to be independent of an individual's sensitivity toward noise. Furthermore, insofar as concentration is concerned, the non-sensitive participants tended to be unaffected, while the sensitive ones were marginally affected; the mean scores for the two groups were 3.01 and 2.63, respectively, with p < 0.05.

This work has attempted to study, for the first time, people's perception of the changes in the indoor and outdoor acoustic environments at residences across Indian cities and towns during the last 2 years or so of the COVID-19 pandemic and their effect on mental well-being, WFH productivity aspects, and online education of children enrolled in schools. A national online cross-sectional survey was carried out by sending a link to potential participants through emails and social media, which attracted 942 responses across India. The survey was an effort to allow people to think in different directions to connect their local sound environment with mental and productivity aspects.

The responses suggest that a significant percentage felt that the indoor environment was noticeably noisier during the stricter lockdown in March–May 2020 and its early UL phases, while during the less stringent lockdown in April–May 2021 and the ensuing ULs, the percentage of participants who perceived the indoor environment to be noisier decreased substantially. As anticipated, most participants attributed the changes in the indoor soundscape to an increased usage of home-entertainment, video-calling with relatives, and online meetings with colleagues as well as increased family interaction. Furthermore, the participants perceived a significant reduction in the outdoor noise levels near their residences during the March–May 2020 lockdown as compared to the pre-COVID times, while during the 2021 lockdown, people felt that the levels were comparable with the pre-COVID times. Indeed, people's perception of the dynamic outdoor noise environment during the last 2 years was supported by environmental noise level graphs of a commercial location in the national capital New Delhi as well as by observing a dynamic noise plot on a map of New Delhi showing the day-to-day noise level variation from January 2019 up to April 2022 at ten different locations; see SuppPubmm1.mp4.²⁸ In an overall sense, the changes in the outdoor soundscape did have a positive impact on mental well-being, therefore, indirectly hinting that people appreciated the relatively quieter conditions that persisted for several months even if they was triggered by a pandemic that continues to pose grave threats to public health.

The responses further revealed that during the lockdowns, the quieter outdoor environments in Indian cities increased feelings of peace, happiness, and concentration and marginally reduced annoyance, but in contrast, the home noise environments only slightly decreased concentration and increased annoyance while not significantly affecting peace and happiness. Furthermore, for females, the mental well-being aspects, annoyance, and claustrophobia were likely to be marginally more affected than for males. Similarly, the well-being aspects of younger participants aged between 21 and 30 were more likely to be adversely affected as compared to the older participants aged 41 and above.

The changes in indoor soundscape did result in a noticeable reduction in productivity and forced the participants to spend extra hours each day to finish the assigned tasks during WFH. As a result, only 15% of participants now prefer the WFH model, and statistical analysis revealed that workplace preference was indeed linked to the prevailing indoor noise environment through its effect on productivity aspects. In particular, nearly half the respondents now prefer a hybrid work-mode that offers greater freedom and convenience, and this is especially true for the young workforce within the age-group 21–30. These statistics contrast entirely with the earlier forecasts, in which a large category and number of jobs were predicted to be remote or WFH-based, especially in the information technology (IT) sector.^35,36 The responses also indicated that the indoor soundscape changes adversely affected education aspects and behavioral attributes during online classes. Most parents or family members now prefer in-person or traditional classroom education for their children over the remote or online education mode.

The responses and the environmental noise data bring to light an interesting aspect that suggests that people will most likely not perceive any major changes in the indoor and outdoor soundscapes should minor restrictions on mobility be enforced in the future to control the spread of new COVID-19 variants. This is because, given their past experiences during 2020 and 2021 lockdowns, it is anticipated that the Indian population has adapted to living with and carrying out their daily work schedules remotely in a home soundscape. It seems unlikely that a situation similar to March–May 2020, during which professional and personal life was completely disrupted, will arise again in India. Having stated this, the responses also emphasize the need to design future building or dwelling soundscapes that better support WFH and online classes. Indeed, as suggested by the previous investigators,^19,25,27 a holistic design approach should be adopted wherein due considerations are given to provide a conducive acoustic environment in rooms at residences where one intends to work or study. This includes incorporating design factors such as better noise insulation from outdoor sound sources and neighboring apartments as well as providing optimal sound absorption treatment on walls and floor to control reverberation and enhance speech clarity. Given the present trend that residences are more than just living, recreational, and sleeping places, acoustic comfort must be taken into account because once the pandemic subsides completely, people may still prefer to remotely work from home on multiple days of a week, although occasionally depending upon their circumstances.⁴³ Otherwise, productivity and well-being will be impacted, as suggested by this study.

Finally, given the enormous diversity in the Indian population, future surveys may need to focus on a greater sample size to obtain a more accurate picture of people's perceptions. It may also be desirable to investigate the relation between the area of work (engineering, IT sector, banking and finance, sales and marketing), typology of the performed tasks, and productivity and well-being aspects during WFH in a given indoor soundscape. Perhaps a greater insight may be achieved by surveying the participants on the specific nature of indoor sound sources that increase feelings of annoyance and whether sharing a room disrupts their preferred indoor soundscape.²⁵ 

This work did not receive financial support from any external funding agency. The authors, however, would like to thank all participants for completing the survey and acknowledge the support provided by the Indian Institute of Technology Kanpur (IITK) in terms of the necessary resources and a congenial research environment. The authors also acknowledge the assistance provided by Shubham Kumar and Randhir Kumar in preparing the supplementary files and processing some data. Finally, the authors are grateful to Professor Rajesh Ranjan at the Department of Aerospace Engineering for helpful discussions and to the anonymous reviewers for their constructive criticisms to improve the presentation.