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 2.45, ONNX Runtime 1.25.0	Download
ONNX	w8a16	Universal	QAIRT 2.45, ONNX Runtime 1.25.0	Download
ONNX	w8a8	Universal	QAIRT 2.45, ONNX Runtime 1.25.0	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® X2 Elite	0.088 ms	181 - 181 MB	NPU
YamNet	ONNX	float	Snapdragon® X Elite	0.168 ms	181 - 181 MB	NPU
YamNet	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.182 ms	0 - 26 MB	NPU
YamNet	ONNX	float	Snapdragon® 8 Gen 1 Mobile	0.362 ms	0 - 34 MB	NPU
YamNet	ONNX	float	Qualcomm® QCS8550 (Proxy)	0.219 ms	0 - 17 MB	NPU
YamNet	ONNX	float	Qualcomm® QCS8450	0.362 ms	0 - 34 MB	NPU
YamNet	ONNX	float	Snapdragon® 8 Elite Mobile	0.153 ms	0 - 21 MB	NPU
YamNet	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.154 ms	0 - 17 MB	NPU
YamNet	ONNX	float	Qualcomm® QCS9075	0.332 ms	0 - 53 MB	NPU
YamNet	ONNX	float	Qualcomm® QCS8750	0.153 ms	0 - 21 MB	NPU
YamNet	ONNX	float	Qualcomm® QCS7181	0.168 ms	181 - 181 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® X2 Elite	0.063 ms	212 - 212 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® X Elite	0.096 ms	177 - 177 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	0.115 ms	0 - 34 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	0.222 ms	0 - 37 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCS6490	0.713 ms	0 - 50 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	0.139 ms	0 - 16 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCS8450	0.222 ms	0 - 37 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.103 ms	0 - 21 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® 8 Elite Mobile	0.102 ms	0 - 24 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCS9075	0.218 ms	0 - 49 MB	NPU
YamNet	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	0.171 ms	0 - 22 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCM6690	0.504 ms	0 - 22 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCS7790	0.171 ms	0 - 22 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCS8750	0.102 ms	0 - 24 MB	NPU
YamNet	ONNX	w8a16	Qualcomm® QCS7181	0.096 ms	177 - 177 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® X2 Elite	0.061 ms	181 - 181 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® X Elite	0.095 ms	149 - 149 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.114 ms	0 - 32 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® 8 Gen 1 Mobile	0.215 ms	0 - 36 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCS6490	0.578 ms	0 - 50 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.136 ms	0 - 15 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCS8450	0.215 ms	0 - 36 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCS9075	0.218 ms	0 - 49 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCM6690	0.424 ms	0 - 21 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	0.171 ms	0 - 21 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.097 ms	0 - 20 MB	NPU
YamNet	ONNX	w8a8	Snapdragon® 8 Elite Mobile	0.097 ms	0 - 25 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCS7790	0.171 ms	0 - 21 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCS8750	0.097 ms	0 - 25 MB	NPU
YamNet	ONNX	w8a8	Qualcomm® QCS7181	0.095 ms	149 - 149 MB	NPU
YamNet	QNN_DLC	float	Snapdragon® X2 Elite	0.181 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	float	Snapdragon® X Elite	0.263 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	0.172 ms	0 - 36 MB	NPU
YamNet	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	0.35 ms	0 - 44 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® QCS8275	0.675 ms	0 - 20 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	0.222 ms	0 - 2 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® QCS8450	0.35 ms	0 - 44 MB	NPU
YamNet	QNN_DLC	float	Snapdragon® 8 Elite Mobile	0.145 ms	0 - 19 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® SA7255P	0.675 ms	0 - 20 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® SA8295P	0.558 ms	0 - 19 MB	NPU
YamNet	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.148 ms	0 - 21 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® QCS9075	0.263 ms	0 - 2 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® QCS8750	0.145 ms	0 - 19 MB	NPU
YamNet	QNN_DLC	float	Qualcomm® QCS7181	0.263 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® X2 Elite	0.152 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® X Elite	0.204 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	0.096 ms	0 - 37 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® 8 Gen 1 Mobile	0.216 ms	0 - 42 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS6490	0.63 ms	2 - 4 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS8275	0.424 ms	0 - 25 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	0.135 ms	0 - 1 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS8450	0.216 ms	0 - 42 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.092 ms	0 - 23 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	0.099 ms	0 - 27 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS9075	0.186 ms	0 - 2 MB	NPU
YamNet	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	0.165 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCM6690	0.506 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® SA8295P	0.42 ms	0 - 22 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® SA7255P	0.424 ms	0 - 25 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS7790	0.165 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS8750	0.099 ms	0 - 27 MB	NPU
YamNet	QNN_DLC	w8a16	Qualcomm® QCS7181	0.204 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.157 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® X Elite	0.194 ms	0 - 0 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.107 ms	0 - 36 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® 8 Gen 1 Mobile	0.206 ms	0 - 41 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS6490	0.532 ms	0 - 2 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS8275	0.39 ms	0 - 25 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.138 ms	0 - 1 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS8450	0.206 ms	0 - 41 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS9075	0.183 ms	0 - 2 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® SA7255P	0.39 ms	0 - 25 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCM6690	0.425 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	0.165 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® SA8295P	0.435 ms	0 - 22 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.093 ms	0 - 23 MB	NPU
YamNet	QNN_DLC	w8a8	Snapdragon® 8 Elite Mobile	0.088 ms	0 - 23 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS7790	0.165 ms	0 - 24 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS8750	0.088 ms	0 - 23 MB	NPU
YamNet	QNN_DLC	w8a8	Qualcomm® QCS7181	0.194 ms	0 - 0 MB	NPU
YamNet	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	0.171 ms	0 - 42 MB	NPU
YamNet	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	0.352 ms	0 - 47 MB	NPU
YamNet	TFLITE	float	Qualcomm® QCS8275	0.695 ms	0 - 27 MB	NPU
YamNet	TFLITE	float	Qualcomm® QCS8550 (Proxy)	0.213 ms	0 - 4 MB	NPU
YamNet	TFLITE	float	Qualcomm® SA8775P	1.577 ms	0 - 21 MB	GPU
YamNet	TFLITE	float	Qualcomm® SA8650P	1.577 ms	0 - 21 MB	GPU
YamNet	TFLITE	float	Qualcomm® SA8255P	1.577 ms	0 - 21 MB	GPU
YamNet	TFLITE	float	Qualcomm® QCS8450	0.352 ms	0 - 47 MB	NPU
YamNet	TFLITE	float	Snapdragon® 8 Elite Mobile	0.159 ms	0 - 26 MB	NPU
YamNet	TFLITE	float	Qualcomm® SA7255P	0.695 ms	0 - 27 MB	NPU
YamNet	TFLITE	float	Qualcomm® SA8295P	0.568 ms	0 - 25 MB	NPU
YamNet	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.152 ms	0 - 26 MB	NPU
YamNet	TFLITE	float	Qualcomm® QCS9075	0.271 ms	0 - 10 MB	NPU
YamNet	TFLITE	float	Qualcomm® QCS8750	0.159 ms	0 - 26 MB	NPU
YamNet	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.115 ms	0 - 36 MB	NPU
YamNet	TFLITE	w8a8	Snapdragon® 8 Gen 1 Mobile	0.206 ms	0 - 42 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS6490	0.527 ms	0 - 6 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS8275	0.418 ms	0 - 24 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.133 ms	0 - 8 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® SA8775P	1.618 ms	0 - 20 MB	GPU
YamNet	TFLITE	w8a8	Qualcomm® SA8650P	1.618 ms	0 - 20 MB	GPU
YamNet	TFLITE	w8a8	Qualcomm® SA8255P	1.618 ms	0 - 20 MB	GPU
YamNet	TFLITE	w8a8	Qualcomm® QCS8450	0.206 ms	0 - 42 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS9075	0.193 ms	0 - 6 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® SA7255P	0.418 ms	0 - 24 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCM6690	0.432 ms	0 - 22 MB	NPU
YamNet	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	0.172 ms	0 - 22 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® SA8295P	0.439 ms	0 - 21 MB	NPU
YamNet	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.096 ms	0 - 22 MB	NPU
YamNet	TFLITE	w8a8	Snapdragon® 8 Elite Mobile	0.096 ms	0 - 22 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS7790	0.172 ms	0 - 22 MB	NPU
YamNet	TFLITE	w8a8	Qualcomm® QCS8750	0.096 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