YamNet is a deep neural network model for audio event classification, developed by Google. Built on the lightweight MobileNet architecture and trained on the large-scale AudioSet dataset, YamNet can recognize over 500 audio event classes, including speech, animal sounds, traffic noise, and musical instruments. It extracts features directly from raw audio waveforms and outputs probable sound event predictions.

Thanks to its efficient and compact design, YamNet is highly suitable for edge deployment, such as on smart speakers, mobile devices, and IoT terminals. Its outputs include class scores, audio embeddings, and frame-level predictions, enabling flexible use in various audio AI tasks like sound event detection, audio search, and acoustic scene understanding.

With strong scalability and easy integration, YamNet serves as a fundamental component in building robust audio perception systems.

Source model repository: YamNet

When the user has fine-tuned the source model, the model conversion process must be performed again.

Users can refer to either of the following two methods to complete the model conversion:

• Using AIMO for model conversion: Click Model Conversion Reference in the Performance Reference section on the right to view the conversion steps.

• Using Qualcomm QNN for model conversion: Please refer to the Qualcomm QNN Documentation.

The model performance benchmarks and example code provided by Model Farm are all implemented based on the APLUX AidLite SDK.

For models in .bin format, you can use either of the following two inference engines to run inference on Qualcomm chips:

Inference using APLUX AidLite: please refer to the APLUX AidLite Developer Documentation

Inference using Qualcomm QNN: Please refer to the Qualcomm QNN Documentation

Inference Example Code

The inference example code is implemented using the AidLite SDK.

Click Model & Code to download the model files and the inference code package. The file structure is as follows:

/model_farm_{model_name}_aidlite
    
    |__ models # folder where model files are stored

    |__ python # aidlite python model inference example

    |__ cpp # aidlite cpp model inference example

    |__ README.md