MiniMax

Text-to-speech service implementation using MiniMax T2A API

Overview

MiniMaxTTSService provides high-quality text-to-speech synthesis using MiniMax’s T2A (Text-to-Audio) API with streaming capabilities, emotional voice control, and support for multiple languages. The service offers various models optimized for different use cases, from low-latency to high-definition audio quality.

MiniMax TTS API Reference

Pipecat’s API methods for MiniMax TTS integration

Example Implementation

Complete example with emotional voice settings

MiniMax Documentation

Official MiniMax T2A API documentation

MiniMax Platform

Access voice models and API credentials

Installation

To use MiniMax services, no additional dependencies are required beyond the base installation:

pip install "pipecat-ai"

Prerequisites

MiniMax Account Setup

Before using MiniMax TTS services, you need:

1. MiniMax Account: Sign up at MiniMax Platform
2. API Credentials: Get your API key and Group ID from the platform
3. Voice Selection: Choose from available voice models and emotional settings

Required Environment Variables

• MINIMAX_API_KEY: Your MiniMax API key for authentication
• MINIMAX_GROUP_ID: Your MiniMax group ID

Configuration

MiniMaxHttpTTSService

api_key · str · required

MiniMax API key for authentication.

group_id · str · required

MiniMax Group ID to identify project.

voice_id · str · default: "Calm_Woman"

Voice identifier for synthesis.

model · str · default: "speech-02-turbo"

TTS model name. Options include speech-2.6-hd, speech-2.6-turbo, speech-02-hd, speech-02-turbo, speech-01-hd, speech-01-turbo.

base_url · str · default: "https://api.minimax.io/v1/t2a_v2"

API base URL. Use https://api.minimaxi.chat/v1/t2a_v2 for mainland China or https://api-uw.minimax.io/v1/t2a_v2 for western United States.

aiohttp_session · aiohttp.ClientSession · required

An aiohttp session for HTTP requests.

sample_rate · int · default: "None"

Output audio sample rate in Hz. When None, uses the pipeline’s configured sample rate.

params · InputParams · default: "None"

Runtime-configurable voice and generation settings. See InputParams below.

InputParams

Voice and generation settings that can be set at initialization via the params constructor argument.

Parameter	Type	Default	Description
language	Language	Language.EN	Language for TTS generation. Supports 40+ languages. Filipino, Tamil, and Persian require speech-2.6-* models.
speed	float	1.0	Speech speed (0.5 to 2.0).
volume	float	1.0	Speech volume (0 to 10).
pitch	int	0	Pitch adjustment (-12 to 12).
emotion	str	None	Emotional tone: "happy", "sad", "angry", "fearful", "disgusted", "surprised", "calm", or "fluent".
english_normalization	bool	None	Deprecated. Use text_normalization instead.
text_normalization	bool	None	Enable text normalization (Chinese/English).
latex_read	bool	None	Enable LaTeX formula reading.
exclude_aggregated_audio	bool	None	Whether to exclude aggregated audio in final chunk.

Usage

Basic Setup

import aiohttp
from pipecat.services.minimax import MiniMaxHttpTTSService

async with aiohttp.ClientSession() as session:
    tts = MiniMaxHttpTTSService(
        api_key=os.getenv("MINIMAX_API_KEY"),
        group_id=os.getenv("MINIMAX_GROUP_ID"),
        aiohttp_session=session,
    )

With Voice Customization

import aiohttp
from pipecat.services.minimax import MiniMaxHttpTTSService
from pipecat.transcriptions.language import Language

async with aiohttp.ClientSession() as session:
    tts = MiniMaxHttpTTSService(
        api_key=os.getenv("MINIMAX_API_KEY"),
        group_id=os.getenv("MINIMAX_GROUP_ID"),
        voice_id="Calm_Woman",
        model="speech-02-hd",
        aiohttp_session=session,
        params=MiniMaxHttpTTSService.InputParams(
            language=Language.ZH,
            speed=1.2,
            emotion="happy",
        ),
    )

Notes

• HTTP-based streaming: MiniMax uses an HTTP streaming API, not WebSocket. Audio data is returned in hex-encoded PCM chunks.
• Emotional voice control: The emotion parameter lets you adjust the emotional tone of the voice without changing the voice model itself.
• Model selection: The speech-2.6-* models are the latest and support additional languages (Filipino, Tamil, Persian). Use turbo variants for lower latency or hd variants for higher quality.
• The Python class is named MiniMaxHttpTTSService, not MiniMaxTTSService.