> ## 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 Amazon Polly

> Move Amazon Polly synthesis to Bland HTTP or stream LLM tokens through Bland realtime TTS.

Choose the migration based on when your text becomes available.

* For a complete string and a downloadable response, use [Synthesize Speech (HTTP)](/api-v2/post/tts).
* For partial LLM text and conversational interruptions, use the [Realtime TTS Quickstart](/tts/realtime-quickstart).

## Complete text

Map a Polly `SynthesizeSpeech` request to Bland as follows:

| Amazon Polly               | Bland `POST /v2/tts`                                    |
| -------------------------- | ------------------------------------------------------- |
| AWS credentials and region | Bland API key                                           |
| `Text`                     | `text`                                                  |
| `VoiceId`                  | Bland voice UUID in `voice`                             |
| `OutputFormat: "pcm"`      | `audio.encoding: "pcm_s16le"`, `audio.container: "raw"` |
| `OutputFormat: "mp3"`      | No native equivalent; request WAV or raw audio          |
| `SampleRate` string        | `audio.sample_rate` number                              |

```bash theme={null}
curl -X POST "https://api.bland.ai/v2/tts" \
  -H "Authorization: Bearer $BLAND_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello from Bland.",
    "voice": "29158307-9893-4149-8a75-bc9ce313d64e",
    "audio": {
      "encoding": "pcm_s16le",
      "sample_rate": 24000,
      "container": "wav"
    }
  }' \
  --output speech.wav
```

Polly SSML, engine selection, lexicons, and speech marks do not map directly. Bland supports pause markers and normalized expressiveness and stability controls on its HTTP endpoint.

## Realtime LLM input

Polly's standard synthesis request takes one complete `Text` value. Bland's WebSocket can start speaking while the LLM is still producing tokens:

```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 }));
```

See the [Amazon Polly synthesis example](https://docs.aws.amazon.com/polly/latest/dg/synthesize-example.html), [Bland HTTP reference](/api-v2/post/tts), and [Bland WebSocket reference](/api-v2/post/tts-ws).

***

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