> ## Documentation Index
> Fetch the complete documentation index at: https://docs.bland.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Migrate from Google Cloud TTS

> Replace Google Cloud bidirectional speech synthesis with Bland's realtime WebSocket.

Google Cloud bidirectional streaming sends one configuration request followed by text input requests over gRPC. Bland uses the same high-level flow over a standard WebSocket.

Start with the [Realtime TTS Quickstart](/tts/realtime-quickstart).

## Request mapping

| Google Cloud streaming TTS                  | Bland `/v2/tts/ws`                                       |
| ------------------------------------------- | -------------------------------------------------------- |
| Google Cloud credentials and project        | Bland API key                                            |
| First `StreamingSynthesisConfig` request    | First `init` message                                     |
| `VoiceSelectionParams.name`                 | Bland voice UUID                                         |
| `StreamingSynthesisInput.text`              | `speak.text` with a turn `context_id`                    |
| End the request iterator                    | Send `end_of_turn`, then keep the socket for later turns |
| `StreamingSynthesizeResponse.audio_content` | Binary WebSocket audio frame                             |

```js theme={null}
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 }));
```

Google's stream ends with the RPC. Bland is designed to stay open across conversational turns. Wait for `utterance_end`, then reuse the same socket with a new `context_id`.

The Google voice name and language code do not carry over. Select a Bland voice UUID that supports your target language. Configure raw PCM or 8 kHz mu-law during `init` and update your playback pipeline for raw binary WebSocket frames.

If your application uses Google's synchronous `SynthesizeSpeech` with complete text, [Synthesize Speech (HTTP)](/api-v2/post/tts) is the closer migration target.

See the [Google bidirectional streaming quickstart](https://docs.cloud.google.com/text-to-speech/docs/create-audio-text-streaming) and [Bland WebSocket reference](/api-v2/post/tts-ws).

***

Docs for agents: [llms.txt](/llms.txt)