Rime TTS (Coda, Mist v3)

Overview

Best for: US contact-centre and telephony agents that need natural conversational English, mu-law output and an on-prem path.

At a glance

Figures are Coda ($50/1M, 8 languages); Mist v3 is $30/1M with 4 languages and TTFB well below 100 ms. Cloud claim sub-200 ms end to end. Concurrency 20 on Starter. BAA and SOC 2 reports on Enterprise. Free allowance unclear (800 vs 3,000 minutes).

Features

• input streaming (/ws3)
• word timestamps
• contextId echo
• flush / clear / eos operations
• segment modes (never, bySentence, immediate)
• on-prem Docker/Kubernetes
• MCP server

Pricing

Free tier: Page conflicts: '~800 minutes free (about 800k characters)' vs FAQ '3,000 free minutes'

Source: rime.ai

Setup

1. Create an API key at rime.ai.
2. Connect to wss://users-ws.rime.ai/ws3 with speaker, modelId and audioFormat query params and an Authorization: Bearer header.
3. Send {text} messages, then {operation:'flush'} or {operation:'eos'}.
4. Decode base64 'chunk' events; use 'timestamps' for barge-in.

Endpoint

wss://users-ws.rime.ai/ws3

Authentication

Authorization: Bearer <RIME_API_KEY> header (browsers cannot set it; proxy via backend)

// npm i ws
import WebSocket from "ws";
import fs from "fs";

const url = "wss://users-ws.rime.ai/ws3?speaker=YOUR_SPEAKER&modelId=mistv3&audioFormat=pcm";
const ws = new WebSocket(url, { headers: { Authorization: `Bearer ${process.env.RIME_API_KEY}` } });
const out = fs.createWriteStream("out.pcm");

ws.on("open", () => {
  ws.send(JSON.stringify({ text: "Hello there. ", contextId: "turn-1" }));
  ws.send(JSON.stringify({ text: "Thanks for calling today." }));
  ws.send(JSON.stringify({ operation: "eos" }));   // speak the rest, then close
});
ws.on("message", (raw) => {
  const m = JSON.parse(raw.toString());
  if (m.type === "chunk") out.write(Buffer.from(m.data, "base64"));
  if (m.type === "error") console.error(m);
});
ws.on("close", () => out.end());

Written from the current docs. Check the vendor's SDK version before you ship.

Warnings

Plus 12 warnings that apply to all text-to-speech, streaming APIs. See category warnings.

Limits

• Starter: 20 concurrent TTS generations
• Rime keeps only one active contextId at a time per socket

Models and products

Docs and sources

Docs

• WebSockets overview
• Changelog
• JSON WebSocket reference

Sources used

• rime.ai/pricing
• docs.rime.ai/docs/websockets
• docs.rime.ai/docs/changelog
• docs.rime.ai/docs/latency