The Rime CLI: From zero to AI voices in seconds
The Rime CLI lets you synthesize speech, preview voices, and generate production-ready API calls directly from your terminal. No dashboard. No context switching. Just your keyboard and 94+ voices that don't sound like robots.
Why a CLI?
The best voice experiences get built by developers who can iterate fast. Switching to a browser to test a voice, copy a curl command, or verify output format slows that loop down. The CLI closes it. You can go from install to audio in under 60 seconds, and stay in flow while you build.
It also fits naturally into the way AI-assisted development is evolving. If you're using Claude Code, you can invoke rime tts directly from your agentic workflow, previewing voices, generating test audio, or scaffolding API calls without ever breaking out of your coding session.
Getting Started
Install
curl -fsSL https://rime.ai/install-cli.sh | sh
Verify it worked:
rime --version
Log in
rime login
This opens your browser once to authenticate. The CLI saves your API key to ~/.rime/rime.toml automatically — you won't need to touch it again.
Hear your first voice
rime hello
That's it. You're talking to Rime.
Synthesize your own text
rime tts "Hello from Rime." -s celeste -m arcana
Want to save the output instead of playing it back?
rime tts "Hey, how's it going?" -s celeste -m arcana -o welcome.wav
Play it back with waveform visualization:
rime play welcome.wav
Try different voices
Swap the --speaker flag to explore voices. We have 94+:
rime tts "The quick brown fox." -s orion -m arcana
rime tts "The quick brown fox." -s luna -m arcana
Generate a curl command for API integration
When you're ready to wire Rime into your app, generate a ready-to-use curl command:
rime curl "Hello from Rime" -s astra -m arcana
The CLI outputs a complete, runnable curl call with your auth header. Add --show-key to include your actual API key, or --oneline for easy copy-paste into scripts.
What's next
The CLI is the fastest path to production-ready voice. Check out the full CLI reference for all commands and flags, browse the voice library to choose the right model for your use case.
