About Elevenlabs CLI
Elevenlabs CLI is an MCP server in the Gaming category: community-built CLI for the ElevenLabs AI audio platform with TTS, STT, voice tools, and MCP server mode.
Install
git clone https://github.com/hongkongkiwi/elevenlabs-cliThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Ask AI
Ask AI about Elevenlabs CLI
Powered by Claude · Grounded in docs
Security checks
- ✓README presentThe repository ships documentation.
5 checks have not been run yet.
README
[!WARNING] This is an independent, community-built CLI and is not an official ElevenLabs release. For official platform docs, visit elevenlabs.io/docs.
Why ElevenLabs CLI
- Unified interface for TTS, STT, voice cloning, dubbing, and audio tooling
- Script-friendly output (
--json) for automation and CI pipelines - MCP server mode for AI assistants with tool filtering and safety controls
- Multi-channel distribution (Homebrew, Scoop, Cargo, Docker, source)
Installation
| Channel | Command |
|---|---|
| Homebrew (macOS/Linux) | brew tap hongkongkiwi/elevenlabs-cli && brew install elevenlabs-cli |
| Scoop (Windows) | scoop bucket add elevenlabs-cli https://github.com/hongkongkiwi/scoop-elevenlabs-cli && scoop install elevenlabs-cli |
| Cargo (all platforms) | cargo install elevenlabs-cli |
| Cargo with MCP feature | cargo install elevenlabs-cli --features mcp |
| Docker | docker run --rm -e ELEVENLABS_API_KEY=your-key ghcr.io/hongkongkiwi/elevenlabs-cli tts "Hello" |
| From source | git clone https://github.com/hongkongkiwi/elevenlabs-cli.git && cd elevenlabs-cli && cargo install --path . |
Quick Start
# 1) Set API key (or use: elevenlabs-cli config set api_key "..." )
export ELEVENLABS_API_KEY="your-api-key"
# 2) Generate speech
elevenlabs-cli tts "Hello from ElevenLabs CLI" --output hello.mp3
# 3) List voices
elevenlabs-cli voice list
# 4) Transcribe audio
elevenlabs-cli stt audio.mp3
[!TIP] Examples use
elevenlabs-cli(the default binary name). If you want a shorter command, add an alias such asalias elevenlabs='elevenlabs-cli'.
Feature Surface
| Area | Commands |
|---|---|
| Speech | tts, tts-stream, tts-timestamps, realtime-tts, stt |
| Voice | voice, voice-changer, voice-design, library, samples, pronunciation |
| Content | sfx, dialogue, music, dub, isolate, audio-native |
| Agent Platform | agent, converse, tools, projects, knowledge, rag, workspace |
| Platform Ops | history, usage, models, user, config, webhook, update, interactive |
| Developer UX | completions, global --json, optional mcp mode |
Common Workflows
Text to Speech
elevenlabs-cli tts "Ship high-quality audio from the terminal" --voice Brian --output narration.mp3
Speech to Text with Speaker Diarization
elevenlabs-cli stt meeting.mp3 --diarize --num-speakers 3
Voice Cloning
elevenlabs-cli voice clone --name "My Voice" --samples voice1.mp3,voice2.mp3
Sound Effects
elevenlabs-cli sfx "A soft UI notification" --duration 3 --output notification.mp3
JSON for Automation
elevenlabs-cli --json voice list | jq '.[0].voice_id'
MCP Server Mode
Start the server:
elevenlabs-cli mcp
[!NOTE]
mcpis available in builds that include themcpfeature (for example,cargo install elevenlabs-cli --features mcp).
Minimal client config:
{
"mcpServers": {
"elevenlabs": {
"command": "elevenlabs-cli",
"args": ["mcp"],
"env": {
"ELEVENLABS_API_KEY": "your-api-key"
}
}
}
}
Security options:
elevenlabs-cli mcp --enable-tools tts,stt,voice
elevenlabs-cli mcp --disable-admin
elevenlabs-cli mcp --disable-destructive
elevenlabs-cli mcp --read-only
Configuration
Default config path: ~/.config/elevenlabs-cli/config.toml
api_key = "your-api-key"
default_voice = "Brian"
default_model = "eleven_multilingual_v2"
default_output_format = "mp3_44100_128"
[mcp]
enable_tools = "tts,stt"
disable_tools = "agents"
disable_admin = false
read_only = false
Available Models
| Model | Best For |
|---|---|
eleven_multilingual_v2 |
High quality synthesis across many languages |
eleven_flash_v2_5 |
Lowest latency |
eleven_turbo_v2_5 |
Balanced speed and quality |
eleven_v3 |
Expressive and emotional speech |
scribe_v1 |
Higher-accuracy speech-to-text |
scribe_v1_base |
Faster, lower-cost speech-to-text |
Output Formats
| Format | Use Case |
|---|---|
mp3_44100_128 |
Default, broadly compatible |
mp3_44100_192 |
Higher quality MP3 |
wav_44100 |
Editing and mastering |
pcm_16000 |
Telephony and real-time systems |
opus_48000_128 |
Streaming and WebRTC |
ulaw_8000 |
Legacy telephony |
Command Reference
Use elevenlabs-cli --help and elevenlabs-cli <command> --help for details.
| Command | Description |
|---|---|
tts |
Text-to-speech synthesis |
stt |
Speech-to-text transcription |
voice |
Voice management (list, clone, edit, delete) |
library |
Shared/community voice library |
isolate |
Background-noise removal |
voice-changer |
Speech-to-speech voice transformation |
dub |
Dubbing and translation workflows |
dialogue |
Multi-speaker dialogue generation |
sfx |
Sound effect generation |
music |
Music generation |
agent, converse, tools, projects |
Conversational/agent features |
knowledge, rag, workspace |
Knowledge and workspace management |
history, usage, models, user |
Account/model/runtime information |
config, webhook, update, interactive, completions |
Tooling and operations |
mcp |
Model Context Protocol server mode |
Ecosystem
Companion repositories and their README files:
| Repository | Purpose | README |
|---|---|---|
| hongkongkiwi/action-elevenlabs-cli | GitHub Action wrapper for CI workflows | Open |
| hongkongkiwi/homebrew-elevenlabs-cli | Homebrew tap for releases | Open |
| hongkongkiwi/scoop-elevenlabs-cli | Scoop bucket for Windows installs | Open |
| hongkongkiwi/elevenlabs-cli-skill | AI-agent skill integration | Open |
Resources
Support
Contributing
Contributions are welcome.
- Fork the repository.
- Create a feature branch.
- Run tests and checks.
- Open a pull request with context and examples.
License
MIT License. See LICENSE.
README copy captured when this server was indexed; it may lag the repository. The original is authoritative.