Voogle is an audio search engine that uses vocal imitations of the desired sound as the search query.
Voogle backend dependencies are installed with
pip install -r requirements.txt.
Note: Windows and Linux users must have FFmpeg installed.
Voogle frontend dependencies are installed with
Note: You must have Node.js installed before you can run
Any collection of audio files can be used as sounds returned by Voogle in response to a vocal query. The Interactive Audio Lab has released 2 datasets specifically for the training of query-by-vocal-imitation models: Vocal Imitation Set and VocalSketch [1, 2]. A small test dataset for demos can be downloaded here.
Interactive Audio Lab has released the following models for query-by-vocal-imitation:
siamese-style: a siamese-style neural network 
VGGish-embedding: cosine similarity of VGGish embeddings 
mcft: multi-resolution common-fate transform 
After installing the dependencies, a dataset, and a model, the Voogle app can be deployed.
- Start the server by running
npm run production.
- Navigate to
localhost:5000in your browser.
From there, please follow the directions found under "Show Instructions". Enjoy!
Note: There are currently two frontend interfaces available for Voogle. If you would like to use the alternate interface, use the command
npm run old-interface instead during step 1.
Unit tests can be run with
npm run test.
Voogle can be extended to incorporate additional models and datasets. If you would like to make your model or dataset available to all users of Voogle, contact firstname.lastname@example.org.
Adding a model
- Define your model as a subclass of
QueryByVoiceModelwith all abstract methods implemented as described.
- Add the model constructor to
- Place your model's weights in
- Update the model name and filepath in
An example model can be found here.
Adding a dataset
- Define your dataset as a subclass of
QueryByVoiceDatasetwith all abstract methods implemented as described.
- Add the dataset constructor to
- Place the audio files in
- Update the dataset name in
An example dataset can be found here.
-  Bongjun Kim, Madhav Ghei, Bryan Pardo, and Zhiyao Duan, "Vocal Imitation Set: a dataset of vocally imitated sound events using the AudioSet ontology," Proceedings of the Detection and Classification of Acoustic Scenes and Events 2018 Workshop (DCASE2018), Surrey, UK, Nov. 2018. [paper link]
-  Mark Cartwright and Bryan Pardo, "Vocalsketch: Vocally imitating audio concepts," Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems (ACM), 2015. [paper link]
-  Yichi Zhang, Bryan Pardo, and Zhiyao Duan, "Siamese Style Convolutional Neural Networks for Sound Search by Vocal Imitation," IEEE/ACM Transactions on Audio Speech and Language Processing. [paper link]
-  Bongjun Kim and Bryan Pardo, "Improving Content-based Audio Retrieval by Vocal Imitation Feedback," IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2019.
-  Fatemeh Pishdadian and Bryan Pardo. “Multi-resolution Common Fate Transform,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2018. [paper link]