@mastra/node-audio

Cross-platform audio I/O toolkit for Node.js — play audio, stream microphone input, and record to disk. Designed for use with voice-enabled agents like Mastra.

🎧 Features

• ✅ Play audio from file, URL, or stream
• 🎤 Get microphone stream with sane defaults
• 📼 Record audio to disk using ffmpeg
• 🔄 Unified createHuddle API for mic/speaker coordination
• 🧰 Built on node-mic, @mastra/node-speaker, and fluent-ffmpeg

🚀 Getting Started

1. Install

npm install @mastra/node-audio

2. System Requirements

This package wraps native audio libraries. You may need additional system dependencies:

🖥 Speaker

• Docs: node-speaker - Audio selection

🎙 Microphone Input

• Docs: node-mic

💾 Recording Audio

• Docs: fluent-ffmpeg
• You must have ffmpeg installed and available in your system path

# macOS
brew install ffmpeg

# Debian/Ubuntu
sudo apt-get install ffmpeg

# Windows
choco install ffmpeg

📚 API Reference

playAudio(input: string | Readable, options?: Speaker.Options): void

Plays a local file, URL, or raw audio stream.

import { playAudio } from "@mastra/node-audio";

playAudio("./response.wav");

You can also pass a stream:

const stream = fs.createReadStream("./hello.wav");
playAudio(stream);

getMicrophoneStream(options?: MicOptions): Readable

Starts recording from the default system mic and returns a stream.

import { getMicrophoneStream } from "@mastra/node-audio";

const micStream = getMicrophoneStream({ rate: 24100 });
micStream.pipe(fs.createWriteStream("raw_input.pcm"));

recordAudioToFile(stream: Readable, outputPath: string, options?: RecorderOptions): Recorder

Records audio from a stream to disk using ffmpeg.

import { getMicrophoneStream, recordAudioToFile } from "@mastra/node-audio";

const mic = getMicrophoneStream();
const recorder = recordAudioToFile(mic, "output.mp3");

// Optionally stop recording later:
// recorder.stream.end();

createHuddle(options: HuddleOptions): Huddle

Unified helper for managing mic + speaker + recording.

import { createHuddle } from "@mastra/node-audio";

const huddle = createHuddle({
  mic: { rate: 24100 },
  speaker: { sampleRate: 24100 },
  record: { outputPath: "session.mp3" },
});

huddle.start();

// Stream mic to agent
const micStream = huddle.getMicrophoneStream();
agent.voice.send(micStream);

// Play response from agent
agent.voice.on("speaker", (stream) => {
  huddle.play(stream);
});

// Gracefully stop
huddle.stop();

🛠 Development

Build both ESM and CommonJS outputs:

npm run build