API cheat sheet

Base URLs and bearer authentication for every Rime API, including speech-to-text and Prism, plus runnable examples for text-to-speech, voice, vocabulary, and text-normalization endpoints.

Create a key on the API Tokens page before running the authenticated examples below.

​

Hosts and authentication

What	Where
TTS over HTTP	https://coda.api.rime.ai/ (Coda), https://mist.api.rime.ai/ (Mist v3)
Coda WebSocket v1	wss://coda.api.rime.ai/ws
Speech-to-text	wss://stt.api.rime.ai/ws (WebSocket), stt.api.rime.ai:443 (gRPC)
Prism (speech to speech)	wss://prism.api.rime.ai/v1/realtime (WebSocket)
Voices and coverage	https://users.rime.ai
Legacy WebSocket endpoints (/ws, /ws2, /ws3)	wss://users-ws.rime.ai
Text normalization	https://optimize.rime.ai
Docs (this site, plus /llms.txt and per-page .md exports)	https://docs.rime.ai

• Send HTTP TTS requests to the model’s host: coda.api.rime.ai or mist.api.rime.ai. The host selects the model, so the body needs no modelId.
• Connect to Coda WebSocket v1 at wss://coda.api.rime.ai/ws. The legacy /ws, /ws2, and /ws3 endpoints stay on users-ws.rime.ai.
• Authenticate with Authorization: Bearer YOUR_API_KEY. On WebSocket connections, send the same value as a connection header.
• Keep the key on a server. Browser WebSocket objects cannot set the required header, so browser applications need a server-side bridge.

Rime publishes Python rime-api and JavaScript @rimelabs/api for generated TTS message definitions and schemas. These packages cover the standalone engine protocol. Your application still manages connections and authentication. The cloud examples below use standard HTTP and WebSocket clients.

The Coda Protobuf quickstart uses rime-api message definitions with a standard WebSocket client.

​

Synthesize speech (HTTP)

POST https://coda.api.rime.ai/ returns audio bytes in the format named by your Accept header (audio/mpeg, audio/wav, audio/webm;codecs=opus, audio/ogg;codecs=opus, audio/L16, audio/PCMU):

curl -X POST https://coda.api.rime.ai/ \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -H 'Accept: audio/mpeg' \
  --output hello.mp3 \
  -d '{"text": "Hello from Rime.", "speaker": "astra"}'

Full parameters and streaming variants: Coda HTTP reference · Streaming guide

​

Stream Coda speech over WebSocket

Connect to wss://coda.api.rime.ai/ws with rime.v1.json and wait for ready. Send these messages separately to stream a sentence as PCM audio:

{"contextId":"turn-1","start":{"speaker":"lyra","language":"en","audioParameters":{"audioFormat":"audio/pcm","samplingRate":24000}}}
{"contextId":"turn-1","text":"Hello from Coda. "}
{"contextId":"turn-1","end":{}}

Decode base64 audio values as they arrive. Read until done, which leaves the socket open. To interrupt, stop playback and send {"contextId":"turn-1","cancel":{}}.

Use the JSON quickstart for a runnable client and the Coda WebSocket API reference for parameters, events, and binary encoding.

For Mist, use the model-specific WebSocket reference.

​

List voices

Both voice endpoints are public and require no API key.

# Voice names, keyed by modelId then ISO 639-2 language code
curl https://users.rime.ai/data/voices/all-v2.json

# Full metadata per voice (gender, age, dialect, language, flagship flag, …)
curl https://users.rime.ai/data/voices/voice_details.json

Choosing a voice: Voices guide · Reference: List All Voices · List Voice Details

​

Check vocabulary coverage (/oov)

Returns the input words that are not in Rime’s pronunciation dictionary:

curl -X POST https://users.rime.ai/oov \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"text": "hello kubectl blorbify"}'
# → ["kubectl","blorbify"]

Out-of-dictionary words still get a best-effort pronunciation. To control them, see Custom pronunciation. Reference: Vocabulary Coverage

​

Normalize text (/textnorm)

Preview exactly how numbers, dates, and phone numbers will be spoken. Note the host: optimize.rime.ai.

curl -X POST https://optimize.rime.ai/textnorm \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"text": "Call 1-800-444-4141 on 3/4/2026"}'
# → {"normalized":"Call one, eight hundred, four four four, four one four one on march fourth twenty twenty six"}

Guide: Text normalization · Reference: Text Normalization

​

HTTP and legacy WebSocket parameters

This table covers HTTP and the legacy WebSocket endpoints. For Coda WebSocket v1, use language and audioParameters inside start. See the Coda WebSocket parameters for the full list.

Parameter	Values	Notes
speaker	See voices	Set it to pick the voice, and match it to the model and language. Over HTTP on the model hosts, a request without it still works: Coda picks a random voice, and Mist v3 uses a default voice.
modelId (legacy WS)	coda (flagship), mistv3	Legacy WebSocket endpoints only. Set it explicitly. Without it, or with an unrecognized value, Mist v3 serves the request. On the model hosts, the host selects the model.
text	string	Up to 1,000 characters per request (HTTP returns 400 “text is too long” beyond that).
lang	en/eng, es/spa, fr/fra, pt/por, de/ger, it/ita, ja/jpn, ar/ara, hi/hin	Must match the speaker’s language.
audioFormat (WS)	wav, mp3, ogg, webm, pcm, mulaw	HTTP uses the Accept header instead.
segment (WS)	bySentence (default), immediate, never	Controls when synthesis starts. See Segmentation.
samplingRate	8000–96000, default 24000	Values above 24000 are upsampling.

​

Where to go deeper

• API reference index: every endpoint across every model
• WebSocket API overview: endpoint comparison, timestamps, interruption handling
• Build a voice agent: complete apps in Next.js, Vite, Express, plain Node, and FastAPI
• Rime CLI and the hosted MCP server: tooling around this same API