Here we report first ever study on acoustical evaluation of Kanheri Caves located in Sanjay Gandhi National Park, Mumbai (Maharashtra, India). These caves are dated to a period between 2nd century BCE to 7th century CE. In this study we used an ambisonic recorder to capture Impulse Response, which carries acoustic signature of the place. Out of total 109 caves 41 were surveyed in available time. Out of those reverberant environment was noted in 12 caves. Measurements were made only in 3 caves (Cave Nos. 1, 3, 11) which are important. In the beginning we carried out an FFT analysis. We then studied room acoustic parameters like Reverberation Time, Early Decay Time, Clarity, Definition, etc., based on the measurement of Impulse response. Cave No. 3 have high value of reverberation time, compared to other. Therefore it also have lower clarity compared to others. It is properties needs to be compared with similar structures (chaityas) in Maharashtra (at Ajanta, Ellora, Nashik, Junnar, etc.) and elsewhere in India. It is worthwhile to carry out further research in Cave No. 3 with more sophisticated instruments as well as 3D modeling. Since the experiment was performed with receiver at only one position, we also suggest to carry out experiment with receiver at multiple positions and then comparing them.
Archaeoacoustic is relatively a new area in scientific studies which combines the principles & technique of Acoustics and the knowledge from archaeology/history about a place under consideration [
Archaeoacoustical investigation in India started some 30 years ago in 1990s. Many sites like Hulimavu Cave Temple (Bengaluru, Karnataka), Udayagiri Cave (Odisha), Koothabalam of Vadakkunathan Temple (Thrissur, Kerala); Rivona Caves, Tambdi Surla Mahadev Temple (all in Goa) and of many other sites have been studied by some researchers [
They are Located [Coordinates: 19°12′30″N, 72°54′23″E] in Sanjay Gandhi National Park in Borivali, Mumbai. This complex houses 109 caves which were recorded earlier, Exploration in recent years have uncovered evidence of more caves and now it has nearly 160 caves. Chronologically they can be divided in 3 phases – First (2nd–4th cent CE), Second (5th–6th cent CE), Third (7th cent CE) [
During the site visit we interacted with local people and security persons. We discussed about the site. Based on the inputs from them Caves 1, 3, 11 and 76 were selected for further experimental work. Due to time issues we could not visit Cave 76 for measurements.
In a survey of 41 caves reverberant environment were noted in Caves 1, 3, 4, 7, 8, 11, 12, 17, 18, 33, 34, 75 (see
Our methodology involves following steps
Interaction with local peoples, Archaeological Survey of India’s officials (as the site is protected by them) about how this places sounds, are there any earlier mentions of such phenomena and its records etc.
Recording of IR using balloon pop (as per ISO 3382 requirement) and Zoom H3-VR (a First Order Ambisonic recorder). Position of sound source is indicated in
After recording they will be processed using Aurora Plugin in Audacity for Fast Fourier Transform (FFT) and Room Acoustics Parameters like Clarity, Reverberation Time, Early Decay Time, Speech Transmission Index. Adobe Audition 3 will be used for Speech Transmission Index (STI). Aurora Plugin, developed by Prof. Angelo Farina is based on ISO-3382 parameters.
It can handle sound pressure input up to 120 dB SPL. It is able to record 360° audio. Highest resolution supported is up to 24 bit/96 kHz. Supported recording formats are Ambisonic A (Raw format), Ambisonic B (FuMa and AmbiX). Microphone position is detected automatically when recording starts. This recorder has built-in motion sensor to generate the playback sound for the desired direction from the data recorded in every direction. It also supports two 3D methods: Ambisonics and Binaural. In order to minimize reflections, it is recommended that recorder must be placed as far from walls and the floor as possible during recording. Binaural recording is supported only up to “48 k/8 bit” or “48 k/16 bit” [
In FOA four signals which are recorded, referred as A-format (raw format without any editing, post processing, conversion to other formats). They are labeled as based on their location as shown in
EDT: It is based on 0 to 10 dB of initial decay. Influenced by early reflections, it depends upon measuring position of source-receiver and room geometry [
D50: It is evaluated at 500 Hz, 1 KHz, 2 KHz, 4 KHz and should have values >50% for good speech intelligibility [
where h2(t)0-50 ms = RIR measured at some distance preferably 1 m, h2(t)0-∞ is a RIR measured with omnidirectional source.
C80: It is a measure of transparency of musical structures [
C50: It is based on the Haas Effect for speech, i.e., when an acoustical reflection reaches within 50 ms of the direct sound. It improves when strong early reflections are present in a room. Optimal values for C50 are ≥3 dB [
where h(t) = power in first 50 ms and thereafter.
Reverberation Time (RT): According to ISO 3382 measured RT is obtained by extrapolation of a 60 dB line fitted to a decay curve. In reality, it is difficult to achieve Signal to Noise Ratio (SNR) which allows 60 dB decay range, therefore –5 dB to –35 dB decay is used, i.e., T30 [
where V = Volume of room in m3, S = Total surface area in m2, α = Absorption coefficient.
We first carried out FFT analysis using Audacity. Results are as shown in
Sr No. | Site name | Freq (Hz) | Sound level (-dB) | Note |
---|---|---|---|---|
1 | Cave 1 | 28, 41, 89, 111, |
62, 68.9, 56.7, 66, |
A0, E1, F2, A2, |
2 | Cave 3 | 42, 86, 118, 135, 192, |
67.9, 63.7, 67.1, 57.5, 52.8, |
E1, F2, A#2, C3, G3, |
3 | Cave 11 | 31, 42, 87, 137, 192, |
70.1, 65.2, 62.5, 55.6, 50.6, |
B0, F1, F2, C#3, G3, |
EDT | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
---|---|---|---|---|---|---|---|---|---|---|---|
CH1 | 14.023 | 1.789 | 1.873 | 2.28 | 1.501 | 0.616 | 0.618 | 0.334 | 0.291 | 0.137 | |
CH2 | 3.908 | 0.954 | 2.972 | 2.226 | 1.832 | 1.024 | 0.679 | 0.426 | 0.324 | 0.206 | |
CH3 | 12.173 | 0.943 | 1.994 | 2.694 | 1.862 | 1.192 | 0.948 | 0.6 | 0.379 | 0.237 | |
CH4 | 4.831 | 3.454 | 2.762 | 3.222 | 2.477 | 1.555 | 1.173 | 0.65 | 0.413 | 0.248 | |
T10 | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
CH1 | 20.676 | 1.312 | 1.952 | 2.908 | 2.473 | 1.488 | 1.077 | 0.692 | 0.526 | 0.238 | |
CH2 | 5.428 | 4.508 | 3.91 | 3.052 | 2.529 | 1.865 | 1.207 | 0.822 | 0.589 | 0.306 | |
CH3 | 12.967 | 0.439 | 1.561 | 3.102 | 2.7 | 2.089 | 1.443 | 0.996 | 0.653 | 0.389 | |
CH4 | 3.882 | 2.308 | 2.906 | 3.495 | 2.826 | 2.179 | 1.579 | 1.098 | 0.685 | 0.416 | |
C50 | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
CH1 | 1.27 | 1.081 | 1.389 | –1.85 | 3.82 | 8.002 | 6.527 | 9.232 | 9.893 | 15.30 | |
CH2 | 3.054 | 4.287 | –0.12 | –2.68 | 1.2 | 5.63 | 6.774 | 8.211 | 9.214 | 12.89 | |
CH3 | –0.033 | 1.827 | –0.46 | –3.33 | 1.266 | 5.211 | 4.015 | 6.359 | 8.538 | 11.47 | |
CH4 | –0.141 | –5.40 | –3.27 | –5.69 | –1.39 | 2.299 | 3.013 | 6.052 | 7.862 | 10.69 | |
D50 | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
CH1 | 57.26 | 56.19 | 57.93 | 39.51 | 70.67 | 86.32 | 81.8 | 89.34 | 90.70 | 97.13 | |
CH2 | 66.891 | 72.85 | 49.30 | 35.03 | 56.86 | 78.52 | 82.63 | 86.88 | 89.29 | 95.12 | |
CH3 | 49.807 | 60.37 | 47.38 | 31.74 | 57.24 | 76.85 | 71.59 | 81.22 | 87.72 | 93.35 | |
CH4 | 49.19 | 22.38 | 31.98 | 21.24 | 42.07 | 62.93 | 66.68 | 80.11 | 85.94 | 92.15 |
EDT | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
---|---|---|---|---|---|---|---|---|---|---|---|
CH1 | 1.322 | 8.32 | 5.852 | 2.984 | 2.41 | 1.149 | 0.57 | 0.286 | 0.16 | 0.073 | |
CH2 | 11.418 | 9.362 | 6.472 | 3.975 | 2.934 | 1.844 | 1.072 | 0.679 | 0.257 | 0.07 | |
CH3 | 0.901 | 7.277 | 4.973 | 2.863 | 2.058 | 1.003 | 0.584 | 0.268 | 0.161 | 0.051 | |
CH4 | 8.594 | 8.925 | 6.061 | 3.544 | 2.36 | 1.337 | 0.754 | 0.497 | 0.189 | 0.108 | |
T10 | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
CH1 | 10.68 | 6.495 | 3.308 | 2.844 | 1.782 | 1.042 | 0.628 | 0.367 | 0.21 | ||
CH2 | 18.754 | 9.76 | 7.477 | 3.84 | 2.68 | 1.781 | 1.212 | 0.791 | 0.51 | 0.255 | |
CH3 | 10.46 | 7.157 | 3.39 | 2.616 | 1.629 | 1.043 | 0.662 | 0.405 | 0.172 | ||
CH4 | 15.872 | 10.17 | 7.896 | 3.83 | 2.759 | 1.736 | 1.246 | 0.779 | 0.413 | 0.248 | |
C50 | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
CH1 | 4.728 | –4.84 | –9.24 | –3.14 | –1.77 | 2.691 | 6.65 | 9.81 | 12.84 | 17.12 | |
CH2 | –2.018 | –10.0 | –11.8 | –6.07 | –8.02 | –3.56 | 0.524 | 3.733 | 10.31 | 15.55 | |
CH3 | 6.34 | –2.81 | –5.60 | –2.18 | –1.47 | 3.04 | 6.92 | 9.959 | 12.90 | 18.04 | |
CH4 | 0.006 | –8.45 | –7.83 | –2.74 | –1.88 | 1.316 | 4.742 | 7.566 | 12.07 | 15.86 | |
D50 | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
CH1 | 74.811 | 24.71 | 10.64 | 32.69 | 39.95 | 65.01 | 82.22 | 90.54 | 95.05 | 98.09 | |
CH2 | 38.589 | 8.968 | 6.212 | 19.83 | 13.62 | 30.59 | 53.01 | 70.26 | 91.48 | 97.29 | |
CH3 | 81.152 | 34.38 | 21.58 | 37.71 | 41.64 | 66.81 | 83.12 | 90.83 | 95.13 | 98.46 | |
CH4 | 50.037 | 12.49 | 14.16 | 34.75 | 39.36 | 57.52 | 74.87 | 85.09 | 94.16 | 97.47 |
EDT | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
---|---|---|---|---|---|---|---|---|---|---|---|
CH1 | 1.318 | 7.403 | 6.291 | 3.074 | 2.551 | 1.347 | 0.516 | 0.152 | 0.126 | 0.04 | |
CH2 | 4.804 | 8.768 | 6.579 | 3.991 | 3.032 | 2.054 | 1.024 | 0.643 | 0.173 | 0.134 | |
CH3 | 3.129 | 7.347 | 5.36 | 2.887 | 2.25 | 1.227 | 0.353 | 0.16 | 0.144 | 0.033 | |
CH4 | 16.538 | 8.43 | 6.267 | 3.606 | 2.705 | 1.66 | 0.714 | 0.335 | 0.169 | 0.082 | |
T10 | Frq.band [Hz] | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k | |
CH1 | 8.991 | 8.396 | 3.432 | 2.699 | 1.901 | 0.937 | 0.497 | 0.353 | 0.176 | ||
CH2 | 6.943 | 8.539 | 4.319 | 2.715 | 2.083 | 1.161 | 0.836 | 0.429 | 0.264 | ||
CH3 | 8.867 | 8.653 | 3.544 | 2.809 | 1.887 | 0.895 | 0.539 | 0.385 | |||
CH4 | 8.513 | 8.589 | 4.383 | 2.847 | 2.17 | 1.159 | 0.672 | 0.449 | 0.235 | ||
C50 | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
CH1 | 4.081 | –4.67 | –8.81 | –4.41 | –2.49 | 1.786 | 7.469 | 11.91 | 13.69 | 18.40 | |
CH2 | –1.607 | –11.4 | –11.4 | –8.20 | –7.74 | –1.45 | 1.734 | 5.003 | 11.96 | 15.29 | |
CH3 | 3.174 | –4.98 | –6.59 | –3.23 | –2.19 | 1.807 | 9.197 | 11.32 | 12.99 | 20.29 | |
CH4 | –2.24 | –7.76 | –7.72 | –4.64 | –3.22 | 1.725 | 5.396 | 9.207 | 12.36 | 16.15 | |
D50 | Frq.band [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1k | 2k | 4k | 8k | 16k |
CH1 | 71.904 | 25.44 | 11.62 | 26.60 | 36.01 | 60.14 | 84.81 | 93.94 | 95.90 | 98.58 | |
CH2 | 40.851 | 6.742 | 6.778 | 13.14 | 14.39 | 41.75 | 59.85 | 75.99 | 94.02 | 97.19 | |
CH3 | 67.5 | 24.08 | 17.96 | 32.22 | 37.67 | 60.26 | 89.26 | 93.13 | 95.21 | 99.07 | |
CH4 | 37.382 | 14.35 | 14.46 | 25.57 | 32.25 | 59.8 | 77.59 | 89.28 | 94.51 | 97.63 |
Sr No. | Site name | C50 | C80 | D50 | EDT | T10 | T20 | T30 | STIPA | STI MALE | STI FEMALE |
---|---|---|---|---|---|---|---|---|---|---|---|
1 | Cave 3 | 2.232 | 4.46 | 56.36 | 2.843 | 4.538 | 5.078 | 5.145 | 0.65 | 0.649 | 0.684 |
2 | Cave 1 | 3.801 | 66.30 | 2.0455 | 2.632 | 2.3943 | 2.0235 | 0.66 | 0.666 | 0.685 | |
3 | Cave 11 | 2.255 | 55.37 | 2.9354 | 4.9117 | 5.6048 | 4.7698 | 0.54 | 0.543 | 0.58 |
We then calculated IACC (Inter-aural Cross Correlation) as shown in
Freq. [Hz] | 31.5 | 63 | 125 | 250 | 500 | 1000 | 2000 | 4000 | 8000 | 16000 |
---|---|---|---|---|---|---|---|---|---|---|
Cave 1 | 0.99 | 0.564 | 0.683 | 0.648 | 0.527 | 0.756 | 0.303 | 0.278 | 0.29 | 0.516 |
Cave 3 | 0.935 | 0.872 | 0.847 | 0.784 | 0.745 | 0.781 | 0.608 | 0.523 | 0.45 | 0.547 |
Cave 11 | 0.973 | 0.919 | 0.891 | 0.793 | 0.684 | 0.667 | 0.341 | 0.577 | 0.43 | 0.527 |
Inter-aural Cross Correlation (IACC) value denotes the correlation between the two signals arriving at two ears. +1 denotes two identical signals with perfect correlation, 0 for no correlation, −1 perfect correlation with out of phase signals (see
Cave 3 being a chaitya have low values of Room acoustic parameters at frequencies above 4k. Room acoustic values in condense form (
We are thankful to Mumbai Circle ASI and its officials for helping us at the site. We are also thankful to First author’s friends Dr. Sagar Padhye (MBBS) and Jayesh Chachad for their help at site and in conducting measurements.