README.md

---
library_name: pytorch
license: other
tags:
- real_time
- android
pipeline_tag: audio-classification

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yamnet/web-assets/model_demo.png)

# 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](https://github.com/w-hc/torch_audioset).
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/yamnet) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).

Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yamnet/releases/v0.47.0/yamnet-onnx-float.zip)
| ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yamnet/releases/v0.47.0/yamnet-onnx-w8a16.zip)
| ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yamnet/releases/v0.47.0/yamnet-onnx-w8a8.zip)
| QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yamnet/releases/v0.47.0/yamnet-qnn_dlc-float.zip)
| QNN_DLC | w8a16 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yamnet/releases/v0.47.0/yamnet-qnn_dlc-w8a16.zip)
| QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yamnet/releases/v0.47.0/yamnet-qnn_dlc-w8a8.zip)
| TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yamnet/releases/v0.47.0/yamnet-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yamnet/releases/v0.47.0/yamnet-tflite-w8a8.zip)

For more device-specific assets and performance metrics, visit **[YamNet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/yamnet)**.


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/yamnet) 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](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/yamnet) 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 | 0.089 ms | 8 - 8 MB | NPU
| YamNet | ONNX | float | Snapdragon® X Elite | 0.287 ms | 8 - 8 MB | NPU
| YamNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.175 ms | 0 - 38 MB | NPU
| YamNet | ONNX | float | Qualcomm® QCS8550 (Proxy) | 0.22 ms | 0 - 28 MB | NPU
| YamNet | ONNX | float | Qualcomm® QCS9075 | 0.3 ms | 0 - 3 MB | NPU
| YamNet | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.157 ms | 0 - 21 MB | NPU
| YamNet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.148 ms | 0 - 24 MB | NPU
| YamNet | ONNX | w8a16 | Snapdragon® X2 Elite | 0.064 ms | 0 - 0 MB | NPU
| YamNet | ONNX | w8a16 | Snapdragon® X Elite | 0.21 ms | 4 - 4 MB | NPU
| YamNet | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.111 ms | 0 - 37 MB | NPU
| YamNet | ONNX | w8a16 | Qualcomm® QCS6490 | 13.793 ms | 11 - 15 MB | CPU
| YamNet | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 0.138 ms | 0 - 5 MB | NPU
| YamNet | ONNX | w8a16 | Qualcomm® QCS9075 | 0.213 ms | 0 - 3 MB | NPU
| YamNet | ONNX | w8a16 | Qualcomm® QCM6690 | 4.691 ms | 0 - 7 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 | 3.404 ms | 0 - 7 MB | CPU
| YamNet | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.097 ms | 0 - 26 MB | NPU
| YamNet | ONNX | w8a8 | Snapdragon® X2 Elite | 0.057 ms | 0 - 0 MB | NPU
| YamNet | ONNX | w8a8 | Snapdragon® X Elite | 0.215 ms | 4 - 4 MB | NPU
| YamNet | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.109 ms | 0 - 38 MB | NPU
| YamNet | ONNX | w8a8 | Qualcomm® QCS6490 | 1.658 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 | 0.892 ms | 0 - 8 MB | CPU
| YamNet | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.096 ms | 0 - 23 MB | NPU
| YamNet | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.633 ms | 0 - 7 MB | CPU
| YamNet | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.098 ms | 0 - 26 MB | NPU
| YamNet | QNN_DLC | float | Snapdragon® X2 Elite | 0.187 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 | 0.168 ms | 0 - 37 MB | NPU
| YamNet | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 0.665 ms | 0 - 22 MB | NPU
| YamNet | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 0.216 ms | 0 - 1 MB | NPU
| YamNet | QNN_DLC | float | Qualcomm® SA8775P | 0.367 ms | 0 - 22 MB | NPU
| YamNet | QNN_DLC | float | Qualcomm® QCS9075 | 0.262 ms | 2 - 4 MB | NPU
| YamNet | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 0.352 ms | 0 - 39 MB | NPU
| YamNet | QNN_DLC | float | Qualcomm® SA7255P | 0.665 ms | 0 - 22 MB | NPU
| YamNet | QNN_DLC | float | Qualcomm® SA8295P | 0.566 ms | 0 - 20 MB | NPU
| YamNet | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.147 ms | 0 - 21 MB | NPU
| YamNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.144 ms | 0 - 24 MB | NPU
| YamNet | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 0.158 ms | 0 - 0 MB | NPU
| YamNet | QNN_DLC | w8a16 | Snapdragon® X Elite | 0.221 ms | 0 - 0 MB | NPU
| YamNet | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.102 ms | 0 - 35 MB | NPU
| YamNet | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 0.603 ms | 2 - 4 MB | NPU
| YamNet | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 0.421 ms | 0 - 23 MB | NPU
| YamNet | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 0.136 ms | 0 - 8 MB | NPU
| YamNet | QNN_DLC | w8a16 | Qualcomm® SA8775P | 0.276 ms | 0 - 24 MB | NPU
| YamNet | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 0.188 ms | 0 - 2 MB | NPU
| YamNet | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 0.518 ms | 0 - 22 MB | NPU
| YamNet | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 0.214 ms | 0 - 36 MB | NPU
| YamNet | QNN_DLC | w8a16 | Qualcomm® SA7255P | 0.421 ms | 0 - 23 MB | NPU
| YamNet | QNN_DLC | w8a16 | Qualcomm® SA8295P | 0.423 ms | 0 - 21 MB | NPU
| YamNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.105 ms | 0 - 21 MB | NPU
| YamNet | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 0.167 ms | 0 - 22 MB | NPU
| YamNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.091 ms | 0 - 25 MB | NPU
| YamNet | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.186 ms | 0 - 0 MB | NPU
| YamNet | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.209 ms | 0 - 0 MB | NPU
| YamNet | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.114 ms | 0 - 35 MB | NPU
| YamNet | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 0.616 ms | 0 - 2 MB | NPU
| YamNet | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 0.405 ms | 0 - 24 MB | NPU
| YamNet | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.139 ms | 0 - 1 MB | NPU
| YamNet | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.281 ms | 0 - 25 MB | NPU
| YamNet | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.183 ms | 0 - 2 MB | NPU
| YamNet | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 0.431 ms | 0 - 23 MB | NPU
| YamNet | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.212 ms | 0 - 36 MB | NPU
| YamNet | QNN_DLC | w8a8 | Qualcomm® SA7255P | 0.405 ms | 0 - 24 MB | NPU
| YamNet | QNN_DLC | w8a8 | Qualcomm® SA8295P | 0.43 ms | 0 - 21 MB | NPU
| YamNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.104 ms | 0 - 25 MB | NPU
| YamNet | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.161 ms | 0 - 22 MB | NPU
| YamNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.098 ms | 0 - 25 MB | NPU
| YamNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.182 ms | 0 - 44 MB | NPU
| YamNet | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 0.699 ms | 0 - 28 MB | NPU
| YamNet | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 0.218 ms | 0 - 3 MB | NPU
| YamNet | TFLITE | float | Qualcomm® SA8775P | 1.149 ms | 0 - 29 MB | NPU
| YamNet | TFLITE | float | Qualcomm® QCS9075 | 0.262 ms | 0 - 10 MB | NPU
| YamNet | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 0.354 ms | 0 - 46 MB | NPU
| YamNet | TFLITE | float | Qualcomm® SA7255P | 0.699 ms | 0 - 28 MB | NPU
| YamNet | TFLITE | float | Qualcomm® SA8295P | 0.571 ms | 0 - 27 MB | NPU
| YamNet | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.161 ms | 0 - 32 MB | NPU
| YamNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.154 ms | 0 - 30 MB | NPU
| YamNet | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.114 ms | 0 - 35 MB | NPU
| YamNet | TFLITE | w8a8 | Qualcomm® QCS6490 | 0.524 ms | 0 - 6 MB | NPU
| YamNet | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 0.423 ms | 0 - 23 MB | NPU
| YamNet | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.139 ms | 0 - 1 MB | NPU
| YamNet | TFLITE | w8a8 | Qualcomm® SA8775P | 0.289 ms | 0 - 24 MB | NPU
| YamNet | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.188 ms | 0 - 6 MB | NPU
| YamNet | TFLITE | w8a8 | Qualcomm® QCM6690 | 0.433 ms | 0 - 21 MB | NPU
| YamNet | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.223 ms | 0 - 37 MB | NPU
| YamNet | TFLITE | w8a8 | Qualcomm® SA7255P | 0.423 ms | 0 - 23 MB | NPU
| YamNet | TFLITE | w8a8 | Qualcomm® SA8295P | 0.446 ms | 0 - 20 MB | NPU
| YamNet | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.102 ms | 0 - 26 MB | NPU
| YamNet | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.169 ms | 0 - 21 MB | NPU
| YamNet | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.086 ms | 0 - 24 MB | NPU

## License
* The license for the original implementation of YamNet can be found
  [here](https://github.com/w-hc/torch_audioset/blob/master/LICENSE).

## References
* [MobileNets Efficient Convolutional Neural Networks for Mobile Vision Applications](https://arxiv.org/abs/1704.04861)
* [Source Model Implementation](https://github.com/w-hc/torch_audioset)

## Community
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
* For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).