whereismyai
Auricle
Auricle is an audio library providing voice activity detection, turn taking, and barge-in for browser voice agents.
Auricle is a Rust program compiled to WebAssembly that acts as a listener for voice agents in the browser. It combines voice activity detection, turn taking, barge-in, and ducking into a single package with an adapter for any OpenAI-compatible speech-to-text, chat, and text-to-speech API. The engine runs locally in a web worker using custom models like Auricle v1 to ignore background speakers and handle interruptions. Developers can use it via script tags, npm packages for JavaScript or React, or integrate it to manage whole conversations and track session metrics. It offers three levels of integration, from a simple detector replacement to a complete conversation manager.
Auricle at a glance
Key features
- Browser-based voice activity detection
- Turn taking and barge-in management
- OpenAI-compatible API adapter
- Local execution via WebAssembly
- Transcript cleanup and filler removal
- React hooks and component support
- Session metrics and stats tracking
Use it for
- Building browser-based voice assistants
- Handling user barge-in during agent speech
- Filtering out background talkers in a room
- Managing conversational turns and audio ducking
Alternatives to Auricle
- AssemblyAI — Industry-leading speech models deliver high accuracy and low latency for voice applications.
- Human or AI? — Identifies AI-generated voices in real time directly inside the browser.
- Pipecat — Builds real-time voice and multimodal agents with frame-level control.
- Moonshine — Runs entirely in your browser with no audio leaving your device.
- browser-whisper — Transcribes audio and video locally with zero servers involved.
- Speaches — It acts as an Ollama equivalent for running local TTS and STT models.
- WeNet — Achieves state-of-the-art results on public speech datasets.
- FluidAudio — Runs local audio AI models on Apple Neural Engine.
- sherpa-onnx — Runs complex audio AI tasks offline on local devices and embedded hardware.
- FunASR — It supports offline, streaming, and edge deployment with multiple models.
- whisper.cpp — It provides high-performance, offline speech recognition on diverse hardware.
- Speech-to-Speech — Fully modular, swappable components with local or hosted execution options.
All Auricle alternatives