# MiniMax (/en/ai/platform/models/tts/minimax)

> 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).


MiniMax provides text-to-speech (TTS) with multiple voice styles and low-latency streaming output.

## Sample configuration

The following example shows how to configure MiniMax 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).

MiniMax supports managed mode and BYOK. In managed mode, Agora provides access to MiniMax, and you don't need a MiniMax account. For BYOK, store your MiniMax API key in a secret set, and refer to it from the agent. For details, see [Use managed mode](/en/ai/platform/build/managed-mode) and [Use your own provider keys](/en/ai/platform/build/use-your-own-keys).

**Python SDK**

```python
from agora_agent import Agent, MiniMaxTTS, 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(
        MiniMaxTTS(
            key=secret_ref("support_prod", "minimax"),  # omit to use managed mode
            group_id="<your_minimax_group_id>",  # omit to use managed mode
            url="wss://api-uw.minimax.io/ws/v1/t2a_v2",  # omit to use managed mode
            model="speech-2.8-turbo",
            voice_id="English_captivating_female1",
            speed=1.0,
            sample_rate=44100,
        )
    )
)
```

**TypeScript SDK**

```ts
import { Agent, MiniMaxTTS, 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 MiniMaxTTS({
      key: secretRef('support_prod', 'minimax'), // omit to use managed mode
      groupId: '<your_minimax_group_id>', // omit to use managed mode
      url: 'wss://api-uw.minimax.io/ws/v1/t2a_v2', // omit to use managed mode
      model: 'speech-2.8-turbo',
      voiceId: 'English_captivating_female1',
      speed: 1.0,
      sampleRate: 44100,
    }),
  );
```

**Go SDK**

```go
// client is your configured Agora client
speed, sampleRate := 1.0, 44100
agent := agentkit.NewAgent(client).
	WithStt( /* configure your ASR vendor */ ).
	WithLlm( /* configure your LLM vendor */ ).
	WithTts(vendors.NewMiniMaxTTS(vendors.MiniMaxTTSOptions{
		Key:        agentkit.SecretRef("support_prod", "minimax"), // omit to use managed mode
		GroupID:    "<your_minimax_group_id>",                     // omit to use managed mode
		URL:        "wss://api-uw.minimax.io/ws/v1/t2a_v2",        // omit to use managed mode
		Model:      "speech-2.8-turbo",
		VoiceID:    "English_captivating_female1",
		Speed:      &speed,
		SampleRate: &sampleRate,
	}))
```

**REST API**

- **Managed mode**

  Set `credential.mode` to `managed` in the `tts` block. Agora provides access to MiniMax, and you don't need a MiniMax account.

  ```json
  "tts": {
    "credential": { "mode": "managed" },
    "vendor": "minimax",
    "params": {
      "model": "speech-2.8-turbo",
      "voice_setting": {
        "voice_id": "English_captivating_female1",
        "speed": 1.0
      },
      "audio_setting": {
        "sample_rate": 44100
      }
    }
  }
  ```


- **BYOK**

  Set `credential.mode` to `byok` and `credential.fields.key` to a reference to your MiniMax API key in a secret set. Your MiniMax group ID isn't secret. Pass it in `params`. In the following example, the key is stored as `minimax` in the `support_prod` set:

  ```json
  "tts": {
    "credential": {
      "mode": "byok",
      "fields": {
        "key": "$secrets.support_prod.minimax"
      }
    },
    "vendor": "minimax",
    "url": "wss://api-uw.minimax.io/ws/v1/t2a_v2",
    "params": {
      "group_id": "<your_minimax_group_id>",
      "model": "speech-2.8-turbo",
      "voice_setting": {
        "voice_id": "English_captivating_female1",
        "speed": 1.0
      },
      "audio_setting": {
        "sample_rate": 44100
      }
    }
  }
  ```

## Key parameters

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

**tts**

**`vendor` (string) required**

Set to `minimax`.

**`credential` (object) required**

How the agent authenticates with MiniMax. Set `mode` to `managed`, or to `byok` with `fields.key` set to a `$secrets.<set>.<key>` reference to your MiniMax API key.

**`url` (string) optional**

A custom WebSocket endpoint for streaming TTS output, for example `wss://api-uw.minimax.io/ws/v1/t2a_v2`. Not needed in managed mode.

**`params` (object) required**

**`group_id` (string) optional**

The MiniMax group identifier. Required for BYOK.

**`model` (string) required**

The TTS model to use, for example `speech-02-turbo`. For managed mode, use `speech-2.6-turbo` or `speech-2.8-turbo`.

**`voice_setting.voice_id` (string) required**

The voice style identifier to use, for example `English_captivating_female1`.

**`voice_setting.speed` (number) required**

The speaking speed, for example `1.0`.

**`audio_setting.sample_rate` (number) optional**

Possible values: `8000, 16000, 22050, 24000, 32000, 44100`

The audio sampling rate in Hz.

> **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 MiniMax without validation. For a full list of supported options, refer to the [MiniMax documentation](https://www.minimax.io/platform/document/t2a_http).