vad-from-microphone

This example demonstrates how to use Voice Activity Detection (VAD) from a microphone in Flutter with sherpa-onnx.

It supports the following VAD models:

• Silero VAD v4
• Ten VAD

It works on the following platforms:

• Android
• iOS
• Linux
• macOS (both arm64 and x86_64 are supported)
• Windows
• Web

Features

• Configurable VAD parameters (threshold, min silence, min speech, max speech)
• Real-time speech detection with circle indicator (black = silent, red = speech)
• Segment counter
• Support for both Silero VAD and Ten VAD models

How to build

1. Download the model

For Silero VAD:

cd flutter-examples/vad-from-microphone/assets
wget https://github.com/k2-fsa/sherpa-onnx/releases/download/vad-models/silero_vad.onnx
cd ..

For Ten VAD:

cd flutter-examples/vad-from-microphone/assets
wget https://github.com/k2-fsa/sherpa-onnx/releases/download/vad-models/ten-vad.onnx
cd ..

Then generate the asset list:

./generate-asset-list.py

2. Select the model

Edit lib/model_config.dart and set selectedModelIndex:

• 0 = Silero VAD v4 (default)
• 1 = Ten VAD

3. Build the APP

• For Linux

flutter build linux

• For macOS

flutter build macos

• For Windows

flutter build windows

• For Android

flutter build apk --split-per-abi --target-platform android-arm64

• For web

flutter run -d chrome

• For iOS

flutter devices
flutter run -d <device-id> --release

Fix for Linux

sudo apt-get install -y libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev libunwind-dev