Acoustic Sounds for Wellbeing: A Novel Dataset and Baseline Results

Alice Baird; Bjoern Schuller

REVIEW 4 major objections 6 minor 49 references

Acoustic Sounds for Wellbeing: A Novel Dataset and Baseline Results

T0 review · 4 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper introduces a new 88-hour dataset of healing sounds and a 57.4% MFCC baseline for classifying them.

desk verdict The ASW dataset is a plausible first for acoustic wellbeing audio, but the paper's central claim is unverifiable because the data is not actually available, and the 'completely mis-classified' chanting statement conflicts with the paper's own confusion matrix. read the letter →

arxiv 1908.01671 v2 pith:MO6WG5WL submitted 2019-08-05 cs.SD eess.AS

Alice Baird , Bjoern Schuller This is my paper · ORCID

classification cs.SDeess.AS

keywords datasetsoundhealingacousticinstrumentsclassificationMFCCsupportvectormachinewellbeingYouTubeaudio

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper establishes the Acoustic Sounds for Wellbeing (ASW) dataset as a new resource for studying audio that is thought to promote wellbeing. It gathers 88+ hours of YouTube audio into five classes: Gongs, Drumming, Chimes, Singing Bowls, and Chanting. The authors also run a first baseline classification and report that conventional MFCC features with a linear support vector machine reach 57.4% unweighted average recall on the full five-class task. This matters because no comparable dataset was publicly available, and a standard benchmark gives future work on soundscape augmentation and audio generation a point of comparison.

What carries the argument

The carrying object is the ASW dataset itself: 68 video-derived audio files totaling 88:03 hours, split into 5,283 60-second chunks over three independent partitions. The classification baseline uses 20 MFCCs extracted with a standard audio toolkit and a linear-kernel support vector machine, with complexity optimized on a validation split and final results reported on held-out test chunks. A second feature set, 4,096-dimensional deep representations from a pre-trained VGG19 network, is included for comparison and performs worse, which the paper interprets as evidence that the long 60-second chunks suit conventional features better at this stage.

What would settle it

Take a random sample of chunks and have three independent annotators label each by ear; if agreement falls far below the level implied by the baseline, or if a large share of chunks is reassigned, the labels and the 57.4% result would not stand.

Watch

Extended reading notes

Core claim

The central claim is that the ASW dataset, collected from creative-commons YouTube videos and cleaned by brief manual listening, is a viable large-scale resource for acoustic wellbeing research. On that dataset, a linear SVM trained on 20 Mel-frequency cepstral coefficients achieves 57.4% unweighted average recall across all five classes, about 37 percentage points above chance; the same pipeline reaches 75.5% on the two-class gong-versus-singing-bowl task. The chanting class is almost entirely misclassified in the five-class setting, which the authors attribute to its small size and acoustic variability.

Load-bearing premise

The dataset's usefulness depends on the assumption that the YouTube search terms and the authors' brief listening pass correctly labelled every chunk into one of the five classes, with no independent verification.

Editorial extensions

If this is right

Future work on classifying wellbeing audio can compare against a fixed ASW train/validation/test split instead of building ad hoc collections.
The 57.4% unweighted average recall becomes the headline benchmark; later feature sets or models need only beat it on the same partition to show progress.
The dataset is large enough to support conditional audio generation or soundscape augmentation experiments, as the authors propose for future work.
Because chanting hurts overall performance, a dedicated collection effort for that class would likely improve multi-class results.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A testable extension would be to re-annotate a random sample of the 5,283 chunks with several listeners and measure inter-annotator agreement; high agreement would strengthen the dataset's label validity, while low agreement would call the 57.4% baseline into question.
The YouTube search strategy likely tracks a 'sound healing' genre as much as acoustic classes; separating instrument acoustics from recording style could be a follow-up experiment.
Because the paper notes that speech was difficult to remove completely, a future version could quantify how much vocal content remains and test whether classification is driven by voice detection rather than instrument identity.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Dear [Name],

The one thing to know about this paper is that its contribution is the ASW dataset, and the dataset is not actually available anywhere. The text says the full dataset is available to the research community, but footnote 1 says the link will be added upon acceptance. So the central claim is unverifiable as posted.

That said, the paper has real merits. The dataset seems to be the first public collection aimed at acoustic wellbeing sounds, with 88+ hours across five classes (gongs, drumming, chimes, singing bowls, chanting) gathered from YouTube under Creative Commons. The collection and post-processing are described in reasonable detail: they listened through files to remove unsuited audio, chunked into 60-second segments, normalized, and removed silence. The class descriptions cite relevant wellbeing literature, which is useful context. The baseline experiments are standard but sensible: SVM with MFCC and Deep Spectrum features, four class configurations, UAR as the metric because of imbalance. The finding that conventional MFCC beats Deep Spectrum on this task is worth reporting.

The soft spots are mostly around claims outstripping the evidence. The missing link is the big one. Second, label quality is uncertain: the manual selection used only one listener, no inter-annotator agreement, and the chanting class was added from the drumming search results, so its purity is unclear. Third, Section 4 says the chanting class is 'completely mis-classified,' but the confusion matrix in Figure 2(d) shows 18 correct chanting predictions out of 68 test segments (26.5% recall). That is low, but not 'completely mis-classified.' The overstatement should be corrected. Minor: no error bars or repeated runs, but that is acceptable for a dataset baseline.

Who is this for? Researchers working on soundscape design, audio generation, or wellbeing audio. If the dataset is released, this becomes a useful resource paper. As posted, it is a description of a dataset that no one can use yet.

My recommendation: send it to peer review, but make the decision conditional on the dataset actually being available and on fixing the chanting overstatement. The underlying idea is sound, and the paper is honestly written apart from that one confusion-matrix phrasing.

Referee Report

4 major / 6 minor

Summary. The paper introduces the Acoustic Sounds for Wellbeing (ASW) dataset, 88+ hours of YouTube-derived audio in five classes (Gongs, Drumming, Chimes, Singing Bowls, Chanting), together with baseline classification experiments using MFCC and Deep Spectrum features with a linear SVM. The authors report unweighted average recall (UAR) up to 57.4% for the full five-class task, noting that the chanting class is difficult. The central claims are that the dataset is a novel resource for sound-healing research and that the baselines provide a reference for future work.

Significance. If the dataset is made publicly available and the reported numbers are verified, the ASW dataset could be a useful community resource for soundscape analysis, audio generation, and wellbeing-related audio research. The baseline classification results, while modest, provide a reasonable starting point for future methods. The paper's use of a held-out test partition is appropriate. However, the contribution is currently not verifiable because the dataset itself is not accessible, and several internal inconsistencies in the data description prevent the reader from trusting the reported statistics.

major comments (4)

[Section 2, footnote 1] The dataset is the central deliverable of this paper, but it is not available: footnote 1 says 'link will be added upon acceptance' and the posted version provides no URL or repository. Without public access, the claimed 68 clips, 5,283 segments, 88:03 total duration, and the classification results cannot be independently confirmed or used by the community. A working download link and a detailed description of the release (file formats, directory structure, metadata, licenses) must be provided before the dataset claim can be accepted.
[Table 1] Table 1 contains an internal inconsistency that undermines the data description. The class rows sum to 34+35+35+34+15 = 153, while the total number of videos is stated as 68 in the '# videos' row; the column sums of the class rows (55, 49, 49) also do not match the partition sizes (24, 22, 22). Either the row/column meanings are mislabeled, or the entries are for a different quantity (e.g., number of segments), but as printed the table is self-contradictory. This needs to be corrected and clarified.
[Section 4, Figure 2(d)] The text states that 'for the 5-class classification (experiment 4), the chanting (cha) class is completely mis-classified,' but the confusion matrix in Figure 2(d) shows 18 correct predictions for the cha row, corresponding to a per-class recall of 18/68 ≈ 26.5%, which is above the 20% chance level. The claim of 'completely mis-classified' is therefore factually incorrect and should be revised to a more accurate characterization, such as 'poorly classified with recall near chance-level but not zero.'
[Section 2.2] The label quality is load-bearing for the dataset's validity, yet the manual listening step is described only as 'listened over briefly' to check 'quality and presence of any additional audio.' No inter-annotator agreement, explicit inclusion/exclusion criteria, or thresholds (e.g., what counts as 'excessive speech') are provided. Since the chanting class was created from drumming search results and the paper reports that speech could not be fully removed, the authors should provide a clearer annotation protocol or at least a quantitative estimate of label purity to support the dataset's use as ground truth.

minor comments (6)

[Abstract] The abstract says '5-classes of acoustic instrumentation (Gongs, Drumming, Singing Bowls, and Chanting)' but lists only four classes; Chimes is missing and should be added.
[Section 3.3 and Table 2] The 'Dev.' column in Table 2 is not clearly defined; the text says complexity is selected on the validation set, but it is not stated whether Dev. reports validation performance of the selected model before retraining on the concatenated train+validation sets. The large gaps between Dev. and Test values (e.g., 34.8% vs. 80.5% in experiment 2) warrant an explicit explanation.
[Section 3.2] The Deep Spectrum extraction is insufficiently specified for reproduction: the paper should state the exact version of the toolkit, the input representation details (e.g., spectrogram window size and hop length), and whether the VGG19 network was used with ImageNet weights or with any fine-tuning.
[Figure 2] The confusion matrices are not annotated with axis labels; the authors should indicate which axis corresponds to true class and which to predicted class.
[Section 2.1] The crawling date, the total number of videos downloaded, and the number of videos discarded at each post-processing step are not reported. Providing these numbers would help assess the data collection process.
[Throughout] There are minor language and typographical errors, including 'significant affect' in the introduction (should be 'effect'), 'affective' in the conclusion (should be 'effective'), and 'reptitious' for 'repetitive' in the description of drumming. A careful proofread is recommended.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the baseline accuracies are measured on a held-out test partition and are not fitted to match a target.

full rationale

This paper makes no derived prediction that reduces to its own inputs. The central empirical claim is the 57.4% UAR 5-class SVM result on MFCC features (Section 4, Table 2, Figure 2d). That result is obtained by training on a training partition, selecting the complexity parameter C on a validation partition, and reporting performance on an unseen test partition, as described in Section 3.3 ('Upon selection of the best performing complexity we report final test results on the unseen test partition against the concatenated train and validation sets'). The class labels come from YouTube search terms plus manual listening (Sections 2.1-2.2), and the classification performance is measured against those labels on held-out data; there is no fitted parameter that is renamed as a prediction. The dataset's validity and availability are separate concerns: the footnote 'link will be added upon acceptance' means the dataset is not yet verifiable, but unavailability is an audibility/reproducibility issue, not circular reasoning. The authors' citations of their own prior work (e.g., [11], [12], [38]) are used as background or as a feature-extraction toolkit, not as the load-bearing justification for the baseline result. No step in the paper defines a class, feature, or metric in terms of the target outcome, and no claimed result is equivalent to its input by construction. The honest finding is therefore no significant circularity.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The paper does not derive any theoretical results. Its assumptions are empirical: that the selected sounds are representative of healing practices, that YouTube is a suitable source, and that manual listening is sufficient to ensure label quality. The only fitted parameter is the SVM complexity C, tuned on the validation set.

free parameters (1)

SVM complexity C = 10^-2, 10^-1, 1 (per experiment)
Optimized on the validation set for each classification task; the reported accuracies depend on this choice.

assumptions (3)

domain assumption The selected sounds (gongs, drums, chimes, singing bowls, chanting) are representative of acoustic wellbeing practices.
Used to justify the dataset's purpose; not tested in this paper, only cited to prior literature.
domain assumption YouTube creative commons videos are a sufficient and appropriate source for collecting these audio classes.
The entire dataset is collected from YouTube; this assumption is not evaluated.
domain assumption Brief manual listening is sufficient to remove files that do not fit the class characteristics.
The cleaning step in Section 2.2 is non-automated and not independently verified.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Acoustic Sounds for Wellbeing: A Novel Dataset and Baseline Results." pith.science (2026). https://pith.science/paper/MO6WG5WL

@misc{pith2026190801671,
  author       = {Pith},
  title        = {Pith review of: Acoustic Sounds for Wellbeing: A Novel Dataset and Baseline Results},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MO6WG5WL}},
  note         = {Machine review of arXiv:1908.01671}
}

read the original abstract

The field of sound healing includes ancient practices coming from a broad range of cultures. Across such practices there is a variety of acoustic instrumentation utilised. Practitioners suggest that sound has the ability to target both mental and even physical health issues, e.g., chronic-stress, or joint-pain. Instruments including the Tibetan singing bowl and vocal chanting, are still widely used today. With the noise-floor of modern urban soundscapes continually increasing and known to impact wellbeing, methods to improve this are needed. With that in mind, this study presents the Acoustic Sounds for Wellbeing (ASW) dataset. The ASW dataset is a dataset gathered from YouTube including 88\,+ hrs of audio from 5-classes of acoustic instrumentation (Gongs, Drumming, Singing Bowls, and Chanting). We additionally present initial baseline classification results on the dataset, finding that conventional Mel-Frequency Cepstra coefficient features achieve at best an unweighted average recalled of 57.4 % for a 5-class support vector machine classification paradigm.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 49 canonical work pages

[1]

Acoustic Sounds for Wellbeing: A Novel Dataset and Baseline Results

INTRODUCTION The soundscape is an often overlooked aspect of our environment, and in urban scenarios the soundscape is becoming increasingly chaotic [1],leading to health issues for those in constant contact [2]. Trafﬁc, roadworks, construction, loud conversations, sirens, such a sonic cacophony results in conditions including insomnia, chronic stress, an...

work page Pith review arXiv 1908
[2]

THE ACOUSTIC SOUNDS FOR WELLBEING DATASET The ASW dataset is a dataset collected from YouTube videos which have been licensed under creative commons. The full ASW dataset is available to the research community 1 and has a total of 88 :05 :23 (hh :mm :ss,), from 68 audio clips (extracted from videos), across 5 classes (Gongs, Drumming, Chimes, Singing Bowl...

work page 1971
[3]

EXPERIMENTAL SETTING As a ﬁrst step for analysis of the classes within the ASW dataset we perform a series of baseline classiﬁcation experiments:

work page
[4]

Singing Bowl (sb) - Given the sim- ilarity of these two instruments, such as the common use of materials (i

2 Class: Gong (gon) vs. Singing Bowl (sb) - Given the sim- ilarity of these two instruments, such as the common use of materials (i. e., often precious metal), we choose this combina- tion to take a closer look at the difference in the data sources for these two classes

work page
[5]

3 Class: Chimes (chi), Drum (dru), Gong - Counter to the 2 Class task, we choose this combination due to their intrinsic acoustic difference

work page
[6]

e., most variety) and smallest class

4 Class: Chimes, Drum, Gong, Singing Bowl - In this experi- ment we remove the Chanting class, as this is the only human related audio sample, as well as the nosiest (i. e., most variety) and smallest class

work page
[7]

5 Class: Chanting (cha), Chime, Drum, Gong, Singing Bowl - All classes of the ASW dataset together, should be seen as the ofﬁcial baseline for the ASW dataset. 3.1. Baseline Partitioning To partition the data into train, validation and test, we chose a data- independent schema. Ultimately this led to a slight imbalance be- tween classes due to the varied ...

work page
[8]

We choose unweighted average recall (UAR) as our evaluation metric due to the imbalance between classes

RESULTS AN DISCUSSION Results from the various experimental setups, are give in Table 2. We choose unweighted average recall (UAR) as our evaluation metric due to the imbalance between classes. Overall the MFCC results are stronger for each classiﬁcation scenario on the test set. Although it is negligible, we do see improve- ment from DEEP SPECTRUM on the...

work page

Show all 49 references

[9]

CONCLUSION AND FUTURE WORK In this study, we have presented the ASW dataset, of 88 + hours of acoustic sounds which are known to encourage wellbeing gathered from YouTube. Results from a baseline classiﬁcation experiement, have shown that at best a 5-class classiﬁcation with M...
[10]

ACKNOWLEDGEMENTS This work is funded by the Bavarian State Ministry of Education, Sci- ence and the Arts in the framework of the Centre Digitisation.Bavaria (ZD.B)
[11]

Ecology of sound: the sonic order of urban space,

Rowland Atkinson, “Ecology of sound: the sonic order of urban space,” Urban studies, vol. 44, no. 10, pp. 1905–1917, 2007

1905
[12]

Jian Kang and Brigitte Schulte-Fortkamp, Soundscape and the built environment, CRC press, 2018

2018
[13]

The role of noise in clinical environments with particular reference to mental health care: A narrative review,

Brian Brown, Peter Rutherford, and Paul Crawford, “The role of noise in clinical environments with particular reference to mental health care: A narrative review,” International journal of nursing studies , vol. 52, no. 9, pp. 1514–1524, 2015

2015
[14]

The selection of introduced sounds to improve the soundscape in the public spaces,

Gil-Soo Jang and Chan Kook, “The selection of introduced sounds to improve the soundscape in the public spaces,” Journal of physiological anthropology and applied human science , vol. 24, no. 1, pp. 55–59, 2005

2005
[15]

R Murray Schafer, The soundscape: Our sonic environment and the tuning of the world , Inner Traditions Bear & Co, Vermont, USA, 1993

1993
[16]

Exploring the effect of sound and music on health in hospital settings: A narrative review,

Timothy Onosahwo Iyendo, “Exploring the effect of sound and music on health in hospital settings: A narrative review,”International Journal of Nursing Studies, vol. 63, pp. 82–100, 2016

2016
[17]

Sounds of healing,

Mitchell L Gaynor, “Sounds of healing,” Country Living’s Healthy Living, 1999

1999
[18]

Holistic healing through music & the arts,

Peggy Leyva Conley, “Holistic healing through music & the arts,”SSRN, 2012

2012
[19]

Complementary therapy for addiction:drumming out drugs,

Michael Winkelman, “Complementary therapy for addiction:drumming out drugs,” American journal of public health , vol. 93, no. 4, pp. 647– 651, 2003

2003
[20]

White noise and sleep induction.,

JA Spencer, DJ Moran, A Lee, and D Talbert, “White noise and sleep induction.,” Archives of disease in childhood, vol. 65, no. 1, pp. 135–137, 1990

1990
[21]

The perceived emotion of isolated synthetic audio: The emosynth dataset and results,

Alice Baird, Emilia Parada-Cabaleiro, Cameron Fraser, Simone Hantke, and Bj¨orn Schuller, “The perceived emotion of isolated synthetic audio: The emosynth dataset and results,” in Proceedings of the Audio Mostly 2018, Wrexham, United Kingdom, 2018, pp. 7:1–7:8, ACM

2018
[22]

Stimulation of psychological listener experiences by semi- automatically composed electroacoustic environments,

Emilia Parada-Cabaleiro, Alice E. Baird, Nicholas Cummins, and Bj¨orn Schuller, “Stimulation of psychological listener experiences by semi- automatically composed electroacoustic environments,” in Proceedings ICME 2017, Hong Kong, P. R. China, July 2017, IEEE, pp. 1051–1056, IEEE

2017
[23]

An overview of sound healing practices: Implications for the profession of music therapy,

Barbara J Crowe and Mary Scovel, “An overview of sound healing practices: Implications for the profession of music therapy,” Music Therapy Perspectives, vol. 14, no. 1, pp. 21–29, 1996

1996
[24]

Create music that will open a person’s heart

Antoinette Olivier and Hetta Potgieter, “Create music that will open a person’s heart”: a perspective on emotional and social wellbeing as depicted in three ﬁlms,” Koersjournal, vol. 80, no. 4, pp. 1–8, 2015

2015
[25]

Sacred wellness: Music and healing among indigenous people.,

P M. Cook, “Sacred wellness: Music and healing among indigenous people.,” in Proc. of Annual Conference of the National Association for Music Therapy Western Region, Seattle, W A, USA, March 1995, pp. 891–897, Colorado State University

1995
[26]

Raymond A. R. MacDonald, Gunter Kreutz, and Laura Mitchell, Music, Health, and Well-being, Oxford University Press, 2013

2013
[27]

Music-evoked nostalgia: Affect, memory, and personality.,

Frederick S Barrett, Kevin J Grimm, Richard W Robins, Tim Wild- schut, Constantine Sedikides, and Petr Janata, “Music-evoked nostalgia: Affect, memory, and personality.,” Emotion, vol. 10, no. 3, pp. 390, 2010

2010
[28]

Walter J Freeman III, A neurobiological role of music in social bonding , UC Berkeley, 1998

1998
[29]

Analyses on sound effects for soundscape design of healing garden at psychiatric hospitals-focused psychological and physiological effects,

Deug-Soo Ahn, “Analyses on sound effects for soundscape design of healing garden at psychiatric hospitals-focused psychological and physiological effects,” Journal of the Korean Institute of Landscape Architecture, vol. 43, no. 1, pp. 82–95, 2015

2015
[30]

Deep V oice 3: Scaling Text-to-Speech with Convolutional Sequence Learning,

Wei Ping, Kainan Peng, Andrew Gibiansky, Sercan O. Arik, Ajay Kan- nan, Sharan Narang, Jonathan Raiman, and John Miller, “Deep V oice 3: Scaling Text-to-Speech with Convolutional Sequence Learning,” 2018

2018
[31]

V oice conversion with cyclic recurrent neural network and ﬁne-tuned wavenet vocoder,

Patrick Lumban Tobing, Yi-Chiao Wu, Tomoki Hayashi, Kazuhiro Kobayashi, and Tomoki Toda, “V oice conversion with cyclic recurrent neural network and ﬁne-tuned wavenet vocoder,” in ICASSP 2019- 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASS...

2019
[32]

Emotional speech synthesis for multi-speaker emotional dataset using wavenet vocoder,

Heejin Choi, Sangjun Park, Jinuk Park, and Minsoo Hahn, “Emotional speech synthesis for multi-speaker emotional dataset using wavenet vocoder,” in 2019 IEEE International Conference on Consumer Elec- tronics (ICCE). IEEE, 2019, pp. 1–2

2019
[33]

Learning representations of emo- tional speech with deep convolutional generative adversarial networks,

Jonathan Chang and Stefan Scherer, “Learning representations of emo- tional speech with deep convolutional generative adversarial networks,” in 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2017, pp. 2746–2750

2017
[34]

A review of psychological liter- ature on the health and wellbeing beneﬁts of biophilic design,

Kaitlyn Gillis and Birgitta Gatersleben, “A review of psychological liter- ature on the health and wellbeing beneﬁts of biophilic design,”Buildings, vol. 5, no. 3, pp. 948–963, 2015

2015
[35]

The Paiste Sound Creation Gongs in Music Therapy,

Joseph J. Moreno, “The Paiste Sound Creation Gongs in Music Therapy,” Music Therapy Perspectives, vol. 7, no. 1, pp. 77–80, 01 1989

1989
[36]

Life rhythm as a symphony of oscillatory patterns: electromagnetic energy and sound vibration modulates gene expression for biological signaling and healing,

David Muehsam and Carlo Ventura, “Life rhythm as a symphony of oscillatory patterns: electromagnetic energy and sound vibration modulates gene expression for biological signaling and healing,”Global advances in Health and Medicine , vol. 3, no. 2, pp. 40–55, 2014

2014
[37]

Altered states during shamanic drumming: a phenomenological study,

Anette Kjellgren and Anders Eriksson, “Altered states during shamanic drumming: a phenomenological study,” International Journal of Transpersonal Studies, vol. 29, no. 2, pp. 3, 2010

2010
[38]

The drum is the shaman, the spear guides his voice,

Janet Hoskins, “The drum is the shaman, the spear guides his voice,” Social Science & Medicine , vol. 27, no. 8, pp. 819–828, 1988

1988
[39]

The role of altered states of consciousness in native american healing,

Timothy C Thomason, “The role of altered states of consciousness in native american healing,” Journal of Rural Community Psychology, E13, vol. 1, pp. 1–11, 2010

2010
[40]

Suren Shrestha, How to heal with singing bowls: Traditional Tibetan healing methods, Sentient publications, 2009

2009
[41]

What is sound healing?,

Simon Heather, “What is sound healing?,” Int. J. Heal. Caring, vol. 7, no. 3, pp. 1–11, 2007

2007
[42]

Chanting for beneﬁts soka gakkai and prosperity in brazil,

Suzana Ramos Coutinho Bornholdt, “Chanting for beneﬁts soka gakkai and prosperity in brazil,” (con) textos: revista d’antropologia i investi- gaci´o social, , no. 4, pp. 64–79, 2010

2010
[43]

Examining the effects of meditation techniques on psychosocial functioning,

David B Wolf and Neil Abell, “Examining the effects of meditation techniques on psychosocial functioning,” Research on Social Work Practice, vol. 13, no. 1, pp. 27–42, 2003

2003
[44]

Shamanism and its discontents,

Michael Fobes Brown, “Shamanism and its discontents,” Medical Anthropology Quarterly, vol. 2, no. 2, pp. 102–120, 1988

1988
[45]

Content-based audio classiﬁcation and retrieval by support vector machines,

Guodong Guo and Stan Z Li, “Content-based audio classiﬁcation and retrieval by support vector machines,” IEEE transactions on Neural Networks, vol. 14, no. 1, pp. 209–215, 2003

2003
[46]

Audio recognition in the wild: Static and dynamic classiﬁcation on a real-world database of animal vocalizations,

Felix Weninger and Bj ¨orn Schuller, “Audio recognition in the wild: Static and dynamic classiﬁcation on a real-world database of animal vocalizations,” in 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2011, pp. 337–340

2011
[47]

librosa: Audio and music signal analysis in python,

Brian McFee, Colin Raffel, Dawen Liang, Daniel PW Ellis, Matt McVicar, Eric Battenberg, and Oriol Nieto, “librosa: Audio and music signal analysis in python,” in Proceedings of the 14th python in science conference, TX,US, 2015, pp. 18–25

2015
[48]

Snore Sound Classiﬁcation Using Image-based Deep Spectrum Features,

Shahin Amiriparian, Maurice Gerczuk, Sandra Ottl, Nicholas Cummins, Michael Freitag, Sergey Pugachevskiy, and Bj ¨orn Schuller, “Snore Sound Classiﬁcation Using Image-based Deep Spectrum Features,” in Proc. INTERSPEECH, Stokholm, Sweden, 2017, pp. 3512–3516, ISCA

2017
[49]

Scikit-learn: Machine learning in Python,

Fabian Pedregosa, Ga ¨el Varoquaux, Alexandre Gramfort, Vincent Michel, Bertrand Thirion, Olivier Grisel, Mathieu Blondel, Peter Pret- tenhofer, Ron Weiss, Vincent Dubourg, et al., “Scikit-learn: Machine learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2...

2011

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Acoustic Sounds for Wellbeing: A Novel Dataset and Baseline Results

INTRODUCTION The soundscape is an often overlooked aspect of our environment, and in urban scenarios the soundscape is becoming increasingly chaotic [1],leading to health issues for those in constant contact [2]. Trafﬁc, roadworks, construction, loud conversations, sirens, such a sonic cacophony results in conditions including insomnia, chronic stress, an...

work page Pith review arXiv 1908

[2] [2]

THE ACOUSTIC SOUNDS FOR WELLBEING DATASET The ASW dataset is a dataset collected from YouTube videos which have been licensed under creative commons. The full ASW dataset is available to the research community 1 and has a total of 88 :05 :23 (hh :mm :ss,), from 68 audio clips (extracted from videos), across 5 classes (Gongs, Drumming, Chimes, Singing Bowl...

work page 1971

[3] [3]

EXPERIMENTAL SETTING As a ﬁrst step for analysis of the classes within the ASW dataset we perform a series of baseline classiﬁcation experiments:

work page

[4] [4]

Singing Bowl (sb) - Given the sim- ilarity of these two instruments, such as the common use of materials (i

2 Class: Gong (gon) vs. Singing Bowl (sb) - Given the sim- ilarity of these two instruments, such as the common use of materials (i. e., often precious metal), we choose this combina- tion to take a closer look at the difference in the data sources for these two classes

work page

[5] [5]

3 Class: Chimes (chi), Drum (dru), Gong - Counter to the 2 Class task, we choose this combination due to their intrinsic acoustic difference

work page

[6] [6]

e., most variety) and smallest class

4 Class: Chimes, Drum, Gong, Singing Bowl - In this experi- ment we remove the Chanting class, as this is the only human related audio sample, as well as the nosiest (i. e., most variety) and smallest class

work page

[7] [7]

5 Class: Chanting (cha), Chime, Drum, Gong, Singing Bowl - All classes of the ASW dataset together, should be seen as the ofﬁcial baseline for the ASW dataset. 3.1. Baseline Partitioning To partition the data into train, validation and test, we chose a data- independent schema. Ultimately this led to a slight imbalance be- tween classes due to the varied ...

work page

[8] [8]

We choose unweighted average recall (UAR) as our evaluation metric due to the imbalance between classes

RESULTS AN DISCUSSION Results from the various experimental setups, are give in Table 2. We choose unweighted average recall (UAR) as our evaluation metric due to the imbalance between classes. Overall the MFCC results are stronger for each classiﬁcation scenario on the test set. Although it is negligible, we do see improve- ment from DEEP SPECTRUM on the...

work page