audiomagic/yamnet-onnx

YAMNet (ONNX)

An ONNX export of Google's YAMNet, an audio event classifier trained on AudioSet. It predicts 521 audio event classes and also exposes a 1024-dimensional embedding that is well suited to transfer learning on small, custom sound datasets.

This repository exists so that Audio Magic's "Detect Sounds" tool can fetch the model from a host we control. It is a straight format conversion — no weights were retrained, fine-tuned, pruned or quantized.

Provenance

Upstream model	google/yamnet version 1 (formerly tfhub.dev/google/yamnet/1)
Upstream code	tensorflow/models → research/audioset/yamnet
Upstream licence	Apache 2.0
Source archive sha256	b80da2a1a56926fb0767205051a200dd7b3beaf3ea1ea126c42a53943996e5e0

Changes made to the original work

As required by Apache 2.0 §4(b), the modifications are:

1. The TensorFlow SavedModel was converted to ONNX with tf2onnx (--saved-model, opset 15). No weight values were altered.
2. yamnet_class_map.csv is copied verbatim from the SavedModel's assets/ directory for convenience.

Nothing else was changed. The original NOTICE file is not reproduced because upstream does not ship one.

Inputs and outputs

The mel-spectrogram frontend is baked into the graph, so the model consumes a raw waveform directly — there is no feature extraction to reimplement (and so no opportunity to get it subtly wrong).

Input — waveform, float32, shape [num_samples]: mono PCM at 16 kHz, nominally in [-1.0, 1.0]. Length is dynamic.

Outputs, one row per frame:

Name	Shape	Meaning
output_0	[frames, 521]	Per-class scores over the AudioSet ontology
output_1	[frames, 1024]	Embeddings (the useful part for transfer learning)
output_2	[frames × 96, 64]	Log-mel spectrogram patches

Each frame covers 0.96 s of audio and the window advances 0.48 s, so frame i spans [i × 0.48, i × 0.48 + 0.96] seconds. Input shorter than one patch is zero-padded and still yields one frame. Empirically:

frames = 1 + ceil(max(0, num_samples - 15360) / 7680)

Usage

import numpy as np, onnxruntime as ort

session = ort.InferenceSession("yamnet.onnx")
waveform = np.zeros(16000 * 3, dtype=np.float32)  # 3 s of 16 kHz mono
scores, embeddings, log_mel = session.run(None, {"waveform": waveform})

Verification

The conversion was checked against YAMNet's documented reference behaviour rather than assumed correct:

Input	Top class
1 kHz sine wave	Beep, bleep
Digital silence	Silence (1.00)
White noise	Static / White noise

Frame counts match the published 0.96 s / 0.48 s framing, and outputs were compared numerically against an independent existing conversion across sine, silence, noise, chirp and sub-second inputs.

Licence

Apache 2.0, inherited from the upstream model and code. Copyright 2022 Google LLC. The full licence text is in LICENSE.

The AudioSet ontology and labels are released by Google under CC BY 4.0.

Citation

YAMNet implements the architecture described in:

@inproceedings{hershey2017cnn,
  title     = {{CNN} Architectures for Large-Scale Audio Classification},
  author    = {Hershey, Shawn and Chaudhuri, Sourish and Ellis, Daniel P. W. and
               Gemmeke, Jort F. and Jansen, Aren and Moore, R. Channing and
               Plakal, Manoj and Platt, Devin and Saurous, Rif A. and
               Seybold, Bryan and Slaney, Malcolm and Weiss, Ron J. and
               Wilson, Kevin},
  booktitle = {2017 IEEE International Conference on Acoustics, Speech and
               Signal Processing (ICASSP)},
  year      = {2017}
}

@inproceedings{gemmeke2017audioset,
  title     = {Audio Set: An ontology and human-labeled dataset for audio events},
  author    = {Gemmeke, Jort F. and Ellis, Daniel P. W. and Freedman, Dylan and
               Jansen, Aren and Lawrence, Wade and Moore, R. Channing and
               Plakal, Manoj and Ritter, Marvin},
  booktitle = {Proc. IEEE ICASSP 2017},
  address   = {New Orleans, LA},
  year      = {2017}
}

It uses the MobileNetV1 depthwise-separable convolution architecture:

@article{howard2017mobilenets,
  title   = {{MobileNets}: Efficient Convolutional Neural Networks for Mobile
             Vision Applications},
  author  = {Howard, Andrew G. and Zhu, Menglong and Chen, Bo and
             Kalenichenko, Dmitry and Wang, Weijun and Weyand, Tobias and
             Andreetto, Marco and Adam, Hartwig},
  journal = {arXiv preprint arXiv:1704.04861},
  year    = {2017}
}