Web based audio recognition

This is an example implementation of local, in-browser audio command recognition using web technologies. The code is split up into three parts:

• Audio feature extraction: /audio-features
• Training a neural network: /train-model
• Running inference on live audio input: /inference-demo

For more information, please see the following blog posts:

1. Audio feature extraction
2. Web based training and inference

Note: this is not an officially supported Google product.