The live-translate-mcp server enables real-time English ↔ Mandarin speech translation, using on-device speech-to-text (Whisper) and text-to-speech (Piper), with Claude handling the linguistic translation. ASR and TTS run fully locally; only the translated text is sent to the Anthropic API.
translate_file: Pass an absolute path to a WAV file. The server auto-detects the language, transcribes the speech, translates it via Claude, synthesizes the translated audio with Piper TTS, saves the output as <name>_translated.wav next to the original, plays it back, and returns the original text, translation, and output file path.
translate_speech: Send base64-encoded WAV audio directly and receive the transcription, translation, and synthesized translated audio as base64 WAV — ideal for programmatic workflows.
health_check: Verify that all required dependencies (Whisper model cache, Piper TTS voice files, and espeak-ng) are present and ready before making translation requests.
Automatic language detection: No need to specify the source language — English or Mandarin Chinese is detected automatically from the audio.
Real-time English ↔ Mandarin speech translation for Claude — powered by Whisper, Claude AI, and Piper TTS.
Give Claude the ability to listen, translate, and speak. live-translate-mcp is a Model Context Protocol (MCP) server that adds speech translation as a native tool inside Claude Desktop and Claude Code. Hand it an audio file, and it transcribes, translates, synthesises, and plays the result — entirely on your machine, with Claude handling the translation.
Step | Technology | Where it runs |
Speech → Text | OpenAI Whisper (via | Local |
Text → Translation | Claude (Opus 4.8) | Anthropic API |
Translation → Speech | Piper TTS (ONNX) | Local |
Audio never leaves your machine except for the translated text sent to the Claude API. ASR and TTS run fully on-device.
Related MCP server: macwhisper-mcp-server
translate_fileTranslate a WAV audio file. Pass an absolute path — the server transcribes it, translates the text via Claude, synthesises speech, saves <name>_translated.wav next to the original, and plays it automatically.
Translate /Users/alice/meeting_clip.wavReturns: original text, translation, and the path to the saved output file.
translate_speechTranslate raw audio passed as a base64-encoded WAV string. Returns the transcription, translation, and synthesised audio as base64 WAV — useful for programmatic workflows.
health_checkVerify that all dependencies (Whisper model cache, Piper voice files, espeak-ng) are present and ready before making a translation request.
No installation required. Run it directly with npx:
npx -y live-translate-mcpOr install globally:
npm install -g live-translate-mcpPrerequisites:
Node.js 18+
An Anthropic API key
espeak-ng — for TTS phonemisation (brew install espeak-ng on macOS, apt install espeak-ng on Linux)
The Whisper model (~150 MB) and Piper voice models (~200 MB) download automatically on first use and are cached in ~/.live-translate/.
Add the following to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"live-translate": {
"command": "npx",
"args": ["-y", "live-translate-mcp"],
"env": {
"ANTHROPIC_API_KEY": "your-api-key-here"
}
}
}
}Restart Claude Desktop. You'll see live-translate appear in the tools panel.
claude mcp add live-translate -- npx -y live-translate-mcpThen set your API key in the environment before starting Claude Code, or pass it via the MCP env config.
Once configured, just ask Claude naturally:
"Translate this audio file for me: /Users/alice/recording.wav"
"Use translate_file on /tmp/interview.wav"
"Check if live-translate is ready"
Claude will call the appropriate tool automatically.
Language | ASR | Translation | TTS |
English | ✓ | ✓ | ✓ |
Mandarin Chinese (普通话) | ✓ | ✓ | ✓ |
Language is detected automatically from the audio — no need to specify it.
live-translate-mcp | Cloud speech APIs | |
ASR | Local (Whisper) | Remote |
Translation | Claude API | Remote |
TTS | Local (Piper) | Remote |
Audio privacy | Audio stays on device | Audio uploaded |
Cost | Claude API only | Per-minute pricing |
Offline | Partially (ASR + TTS) | No |
Variable | Default | Description |
| — | Required. Your Anthropic API key. |
|
| Path to Piper |
live-translate — the standalone CLI for push-to-talk translation with no API key required, using local models end-to-end.
Model Context Protocol — the open standard this server implements.
Piper TTS — the local TTS engine powering speech synthesis.
MIT
Check whether all live-translate-mcp dependencies are available: Whisper model cache, Piper voice files, and espeak-ng binary.
Inputs: no named parameters.
Translate a speech audio file between English and Mandarin Chinese. Pass an absolute path to a WAV file. Automatically detects the input language and translates to the other. Saves the translated audio next to the source file and plays it. Returns original text and translation.
Inputs: file_path (required).
Translate speech audio between English and Mandarin Chinese. Automatically detects the input language and translates to the other. Returns original text, translation, and synthesised audio as base64-encoded WAV.
Inputs: audio_base64 (required), sample_rate.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 3 tools
Each tool has a clearly distinct purpose: health_check checks dependencies, translate_file handles file-based translation with local saving, and translate_speech provides real-time speech translation with base64 output. No overlap.
All tool names use consistent snake_case naming with a verb_noun pattern (health_check, translate_file, translate_speech), making it easy to infer functionality.
With 3 tools, the server is small but well-scoped for its purpose of live translation between English and Mandarin. The count is reasonable, though slightly on the lower end, covering health check, file translation, and speech translation.
The server covers the core workflow of speech translation and file conversion but lacks text translation, language configuration, or support for additional language pairs, which are notable gaps for a translation tool.
Claude makes real phone calls for you — in many languages, with transcript and outcome back in chat.
Transcribe audio and video with Speechmatics speech-to-text from Claude and any MCP client.
Audio for your agent: transcribe, speak, translate, summarise, plus sound effects and music.
Talk to your public-facing AI from any MCP client — Claude, ChatGPT, Cursor, Cline, Windsurf.
