# ElevenLabs (/en/ai/platform/models/tts/elevenlabs)

> For AI agents: see the complete documentation index at [llms.txt](/llms.txt).

> Documentation: Conversational AI.
> Documentation version: Preview.
> This document describes only this version. [Version index](/llms/doc-versions-convoai-guides-v2.txt).


ElevenLabs provides highly realistic AI voices with advanced prosody and natural speech patterns, delivering lifelike audio synthesis with emotional nuance and conversational flow.

> **Caution: Paid plan required**
>
> You need a **paid ElevenLabs plan** for reliable TTS integration.
>
> ElevenLabs may restrict or disable free-tier accounts due to abuse-detection mechanisms, even if free credits are available. To avoid missing audio responses during testing and production, ensure you use a paid plan.

## Sample configuration

The following example shows how to configure ElevenLabs as the agent's TTS when you create or update an agent. For a complete example, see [Create an agent and start a session](/en/ai/platform/get-started/start-stop-agent#create-an-agent).

ElevenLabs supports BYOK only. It isn't available in managed mode. Store your ElevenLabs API key in a secret set, and refer to it from the agent. For details, see [Use your own provider keys](/en/ai/platform/build/use-your-own-keys).

**Python SDK**

```python
from agora_agent import Agent, ElevenLabsTTS, secret_ref

# client is your configured Agora client
agent = (
    Agent(client=client)
    .with_stt(...)  # configure your ASR vendor
    .with_llm(...)  # configure your LLM vendor
    .with_tts(
        ElevenLabsTTS(
            key=secret_ref("support_prod", "elevenlabs"),
            base_url="wss://api.elevenlabs.io/v1",
            model_id="eleven_flash_v2_5",
            voice_id="pNInz6obpgDQGcFmaJgB",
            sample_rate=24000,
        )
    )
)
```

**TypeScript SDK**

```ts
import { Agent, ElevenLabsTTS, secretRef } from 'agora-agents';

// client is your configured Agora client
const agent = new Agent({ client })
  .withStt(/* configure your ASR vendor */)
  .withLlm(/* configure your LLM vendor */)
  .withTts(
    new ElevenLabsTTS({
      key: secretRef('support_prod', 'elevenlabs'),
      baseUrl: 'wss://api.elevenlabs.io/v1',
      modelId: 'eleven_flash_v2_5',
      voiceId: 'pNInz6obpgDQGcFmaJgB',
      sampleRate: 24000,
    }),
  );
```

**Go SDK**

```go
// client is your configured Agora client
sampleRate := vendors.SampleRate24kHz
agent := agentkit.NewAgent(client).
	WithStt( /* configure your ASR vendor */ ).
	WithLlm( /* configure your LLM vendor */ ).
	WithTts(vendors.NewElevenLabsTTS(vendors.ElevenLabsTTSOptions{
		Key:        agentkit.SecretRef("support_prod", "elevenlabs"),
		BaseURL:    "wss://api.elevenlabs.io/v1",
		ModelID:    "eleven_flash_v2_5",
		VoiceID:    "pNInz6obpgDQGcFmaJgB",
		SampleRate: &sampleRate,
	}))
```

**REST API**

Set `credential.mode` to `byok` and `credential.fields.key` to a reference to your ElevenLabs API key. In the following example, the key is stored as `elevenlabs` in the `support_prod` set:

```json
"tts": {
  "credential": {
    "mode": "byok",
    "fields": {
      "key": "$secrets.support_prod.elevenlabs"
    }
  },
  "vendor": "elevenlabs",
  "url": "wss://api.elevenlabs.io/v1",
  "params": {
    "model_id": "eleven_flash_v2_5",
    "voice_id": "pNInz6obpgDQGcFmaJgB",
    "sample_rate": 24000
  }
}
```

## Key parameters

The `tts` block takes the following fields. In the SDKs, `ElevenLabsTTS` sets them from its options.

**tts**

**`vendor` (string) required**

Set to `elevenlabs`.

**`credential` (object) required**

Set `mode` to `byok` and `fields.key` to a `$secrets.<set>.<key>` reference to your ElevenLabs API key. Get your API key from the [ElevenLabs Console](https://elevenlabs.io/).

**`url` (string) required**

The endpoint URL for the ElevenLabs TTS service. See [Data residency](https://elevenlabs.io/docs/product-guides/administration/data-residency).

**`params` (object) required**

**`model_id` (string) required**

Identifier of the model to be used. Popular options include `eleven_flash_v2_5` for speed or `eleven_multilingual_v2` for quality.

**`voice_id` (string) required**

The identifier for the selected voice for speech synthesis. Browse available voices in the [Voice Library](https://elevenlabs.io/voice-library).

**`sample_rate` (number) optional**

Default value: `24000`

Audio sampling rate in Hz. Common values: `16000`, `22050`, `24000`, `44100`.

**`speed` (number) optional**

Default value: `1.0`

Speed up or slow down the speed of the generated speech. Range `0.7` to `1.2` inclusive.

**`stability` (number) optional**

Controls voice stability. Higher values `(0.8-1.0)` produce more consistent speech, lower values `(0.0-0.5)` add more variation.

**`similarity_boost` (number) optional**

Enhances similarity to the original voice. Range: `0.0-1.0`. Higher values stick closer to the training voice.

**`style` (number) optional**

Controls speaking style and expressiveness. Higher values increase emotional range and variation.

**`use_speaker_boost` (boolean) optional**

Improves voice quality and similarity when enabled. Recommended for most use cases.

> **Caution**
>
> The parameters listed on this page are validated for use with the Conversational AI Platform. Required parameters must be provided as documented. Any additional parameters in `params` are passed through directly to ElevenLabs without validation. For a full list of supported options, refer to the [ElevenLabs TTS documentation](https://elevenlabs.io/docs/capabilities/text-to-speech).