Readme
MiniMax TTS: Professional Text-to-Speech API
Transform your text into natural-sounding speech with MiniMax TTS, a powerful text-to-speech API designed for developers who need high-quality voice synthesis in their applications.
Overview
MiniMax TTS delivers studio-quality voice synthesis with advanced neural networks, offering natural prosody, clear articulation, and emotion-aware speech generation. Perfect for applications ranging from audiobook production to accessible content creation.
Key Capabilities
Generate lifelike speech with precise control over:
- Voice characteristics and speaking style
- Speech pacing and emotional tone
- Multiple language support with natural accents (30+ languages)
- Real-time speech synthesis for interactive applications
- Over 300+ authentic voices
Getting Started
Getting up and running with MiniMax TTS takes just a few minutes. Here's how to begin:
- Install the client library:
# Using npm
npm install --save @fal-ai/client
# Using pip
pip install fal-client
- Configure your API key:
import { fal } from "@fal-ai/client";
fal.config({
credentials: "YOUR_FAL_KEY_HERE"
});
- Make your first API call:
const result = await fal.subscribe("fal-ai/minimax/speech-02-hd", {
input: {
text: "Welcome to MiniMax TTS. This is a demonstration of our natural speech synthesis."
}
});
Technical Integration
MiniMax TTS supports both synchronous and queue-based processing. The API accepts plain text input and returns audio in multiple formats including MP3, WAV, FLAC, and PCM.
Advanced Usage
Control fine-grained speech parameters:
const response = await fal.subscribe("fal-ai/minimax/speech-02-hd", {
input: {
text: "This is an important announcement.",
voice_setting: {
voice_id: "Wise_Woman",
speed: 1.0,
vol: 1.0,
pitch: 0,
english_normalization: false
},
output_format: "mp3"
}
});
Queue-Based Processing
For handling multiple requests or asynchronous workflows:
// Submit request
const { request_id } = await fal.queue.submit("fal-ai/minimax/speech-02-hd", {
input: {
text: "Your text here"
},
webhookUrl: "https://optional.webhook.url/for/results"
});
// Check status
const status = await fal.queue.status("fal-ai/minimax/speech-02-hd", {
requestId: request_id,
logs: true
});
// Get result
const result = await fal.queue.result("fal-ai/minimax/speech-02-hd", {
requestId: request_id
});
Voice Settings
The API provides extensive control over voice parameters:
{
voice_setting: {
voice_id: string, // One of 300+ available voices
speed: number, // Speed control (default: 1.0)
vol: number, // Volume control (default: 1.0)
pitch: number, // Pitch adjustment (default: 0)
english_normalization: boolean // Improves number reading
}
}
Supported Languages
MiniMax TTS supports 30+ languages including:
- Chinese, Chinese (Yue/Cantonese)
- English, Spanish, French, German, Italian, Portuguese
- Japanese, Korean
- Arabic, Russian, Turkish, Dutch, Ukrainian
- Vietnamese, Indonesian, Thai, Hindi
- Polish, Romanian, Greek, Czech, Finnish
Available audio output formats:
- MP3 - Default format, good compression
- WAV - Uncompressed, high quality
- FLAC - Lossless compression
- PCM - Raw audio data
Best Practices
For optimal results with MiniMax TTS:
- Include proper punctuation in your input text for natural pausing and intonation
- Use the
english_normalization flag for better number reading performance
- Process up to 5,000 characters in real-time or 1 million characters asynchronously
- Cache frequently used audio outputs to optimize performance and costs
Error Handling
Implement robust error handling to manage API responses:
try {
const result = await fal.subscribe("fal-ai/minimax/speech-02-hd", {
input: { text: inputText }
});
} catch (error) {
console.error("Speech generation failed:", error.message);
// Implement fallback behavior
}
Available Models
MiniMax offers multiple TTS models:
- speech-02-hd: High-definition quality, best for production use
- speech-02-turbo: Optimized for real-time applications with low latency
- speech-01-hd: Previous generation HD model
- speech-01-turbo: Previous generation turbo model
MiniMax TTS is built for production workloads with:
- Low latency response times for real-time applications
- High-throughput capability for batch processing
- Automatic scaling to handle varying loads
- Global CDN distribution for consistent performance
Pricing and Usage
- Cost: $0.05 per 1000 characters
- Transparent, usage-based pricing
- No subscription necessary
- No hidden fees or minimum commitments
View detailed pricing or contact sales for enterprise solutions.
Support Resources
We're here to help you succeed:
Start building with MiniMax TTS today and bring natural speech to your applications.
fal-ai/minimax-tts/text-to-speech