Migrate from PlayHT

PlayHT requires an HTTP authentication request to obtain generated, model-specific WebSocket URLs, then accepts one TTS command per generation. Bland uses a stable WebSocket URL and authenticates the upgrade directly. Start with the Realtime TTS Quickstart.

Connection changes

PlayHT	Bland
`POST /api/v4/websocket-auth` with API key and user ID	Connect directly to `wss://api.bland.ai/v2/tts/ws`
Generated URL from `websocket_urls`	Stable WebSocket URL
PlayHT API key plus user ID	Bland API key in `Authorization: Bearer ...`
Voice and output settings in each TTS command	Voice and output settings once in `init`

Request mapping

PlayHT accepts a complete text value for one request and returns start, binary audio, and end. Bland can begin from partial input:

const turnId = crypto.randomUUID();

for await (const token of llmTextStream) {
  ws.send(
    JSON.stringify({
      type: "speak",
      context_id: turnId,
      text: token,
    }),
  );
}

ws.send(JSON.stringify({ type: "end_of_turn", context_id: turnId }));

Map PlayHT start and end handling to Bland utterance_start and utterance_end. Binary audio remains binary, but configure its format during init:

{
  "type": "init",
  "voice": "29158307-9893-4149-8a75-bc9ce313d64e",
  "audio": { "encoding": "pcm_s16le", "sample_rate": 48000 }
}

Bland realtime output supports raw PCM or 8 kHz mu-law. Use Synthesize Speech (HTTP) if your current PlayHT integration needs a complete WAV file. MP3, OGG, FLAC, speed, seed, and PlayHT engine settings do not map directly to the Bland realtime protocol. See the PlayHT WebSocket reference and Bland WebSocket reference.

Docs for agents: llms.txt

⌘I

Get Started

Realtime Speech

HTTP Speech

Migrate

Voice Cloning

Voices

Discovery

Samples

Generation History

Models

Tooling

Migrate from PlayHT

Connection changes

Request mapping

​Connection changes

​Request mapping

Connection changes

Request mapping