YamNet: Optimized for Qualcomm Devices

An audio event classifier trained on the AudioSet dataset to predict audio events from the AudioSet ontology employing the Mobilenet_v1 depthwise-separable convolution architecture.

This is based on the implementation of YamNet found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime	Precision	Chipset	SDK Versions	Download
ONNX	float	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	Download
ONNX	w8a16	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
QNN_DLC	w8a16	Universal	QAIRT 2.45	Download
QNN_DLC	w8a8	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download
TFLITE	w8a8	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit YamNet on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

Custom weights (e.g., fine-tuned checkpoints)
Custom input shapes
Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for YamNet on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.audio_classification

Model Stats:

Model checkpoint: yamnet.pth
Input resolution: 1x1x96x64
Number of parameters: 3.73M
Model size (float): 14.2 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
YamNet	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.146 ms	0 - 24 MB	NPU
YamNet	ONNX	float	Snapdragon® X2 Elite	0.123 ms	8 - 8 MB	NPU
YamNet	ONNX	float	Snapdragon® X Elite	0.277 ms	8 - 8 MB	NPU
YamNet	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.183 ms	0 - 38 MB	NPU
YamNet	ONNX	float	Qualcomm® QCS8550 (Proxy)	0.222 ms	0 - 94 MB	NPU
YamNet	ONNX	float	Qualcomm® QCS9075	0.327 ms	0 - 3 MB	NPU
YamNet	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	0.165 ms	0 - 26 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.103 ms	0 - 27 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® X2 Elite	0.065 ms	0 - 0 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® X Elite	0.222 ms	4 - 4 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	0.109 ms	0 - 39 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCS6490	13.726 ms	13 - 17 MB	CPU
YamNet	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	0.142 ms	0 - 16 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCS9075	0.218 ms	0 - 3 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCM6690	4.651 ms	1 - 8 MB	CPU
YamNet	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.1 ms	0 - 23 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	3.405 ms	0 - 7 MB	CPU
YamNet	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.093 ms	0 - 26 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® X2 Elite	0.091 ms	0 - 0 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® X Elite	0.208 ms	4 - 4 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.115 ms	0 - 38 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCS6490	1.642 ms	0 - 6 MB	CPU
YamNet	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.141 ms	0 - 6 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCS9075	0.218 ms	0 - 3 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCM6690	0.878 ms	0 - 6 MB	CPU
YamNet	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.1 ms	0 - 23 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	0.616 ms	0 - 7 MB	CPU
YamNet	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.14 ms	0 - 23 MB	NPU
YamNet	QNN_DLC	float	Snapdragon® X2 Elite	0.199 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	float	Snapdragon® X Elite	0.273 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	0.18 ms	0 - 36 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	0.671 ms	0 - 21 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	0.217 ms	0 - 23 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® SA8775P	0.361 ms	0 - 22 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® QCS9075	0.26 ms	0 - 2 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	0.353 ms	0 - 38 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® SA7255P	0.671 ms	0 - 21 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® SA8295P	0.574 ms	0 - 20 MB	NPU
YamNet	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	0.157 ms	0 - 20 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.099 ms	0 - 26 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® X2 Elite	0.154 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® X Elite	0.211 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	0.104 ms	0 - 36 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS6490	0.553 ms	0 - 2 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	0.415 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	0.132 ms	0 - 7 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® SA8775P	0.28 ms	0 - 25 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS9075	0.186 ms	2 - 4 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCM6690	0.498 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS8450 (Proxy)	0.211 ms	0 - 37 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® SA7255P	0.415 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® SA8295P	0.432 ms	0 - 22 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.103 ms	0 - 23 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	0.171 ms	0 - 23 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.099 ms	0 - 26 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.158 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® X Elite	0.219 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.113 ms	0 - 36 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS6490	0.625 ms	0 - 2 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	0.391 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.134 ms	0 - 58 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® SA8775P	0.272 ms	0 - 25 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS9075	0.183 ms	0 - 2 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCM6690	0.438 ms	0 - 23 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	0.203 ms	0 - 37 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® SA7255P	0.391 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® SA8295P	0.429 ms	0 - 22 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.093 ms	0 - 22 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	0.166 ms	0 - 23 MB	NPU
YamNet	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.151 ms	0 - 29 MB	NPU
YamNet	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	0.189 ms	0 - 42 MB	NPU
YamNet	TFLITE	float	Qualcomm® QCS8275 (Proxy)	0.683 ms	0 - 27 MB	NPU
YamNet	TFLITE	float	Qualcomm® QCS8550 (Proxy)	0.218 ms	0 - 2 MB	NPU
YamNet	TFLITE	float	Qualcomm® SA8775P	0.378 ms	0 - 28 MB	NPU
YamNet	TFLITE	float	Qualcomm® QCS9075	0.262 ms	0 - 10 MB	NPU
YamNet	TFLITE	float	Qualcomm® QCS8450 (Proxy)	0.352 ms	0 - 44 MB	NPU
YamNet	TFLITE	float	Qualcomm® SA7255P	0.683 ms	0 - 27 MB	NPU
YamNet	TFLITE	float	Qualcomm® SA8295P	0.573 ms	0 - 25 MB	NPU
YamNet	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	0.165 ms	0 - 31 MB	NPU
YamNet	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.094 ms	0 - 25 MB	NPU
YamNet	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.112 ms	0 - 37 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS6490	0.519 ms	0 - 6 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	0.426 ms	0 - 24 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.137 ms	0 - 11 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® SA8775P	0.285 ms	0 - 25 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS9075	0.186 ms	0 - 6 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCM6690	0.431 ms	0 - 22 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	0.204 ms	0 - 38 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® SA7255P	0.426 ms	0 - 24 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® SA8295P	0.428 ms	0 - 21 MB	NPU
YamNet	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.1 ms	0 - 27 MB	NPU
YamNet	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	0.165 ms	0 - 22 MB	NPU

License

The license for the original implementation of YamNet can be found here.

References

Community

Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
For questions or feedback please reach out to us.

Downloads last month: -; Downloads are not tracked for this model. How to track

Paper for qualcomm/YamNet

MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

Paper • 1704.04861 • Published Apr 17, 2017 • 1