minimax-tts

Use MiniMax speech-2.8-hd model for high-quality text-to-speech synthesis. Supports multiple Chinese and English voices. Install when needed.

MiniMax Text-to-Speech

Call via: python scripts/minimax_media.py tts "<text>" [options]

Usage

python scripts/minimax_media.py tts "你好，這是語音測試。" --voice female-tianmei --speed 1.0

Options

• --voice: Voice ID (see below)
• --speed: Speech speed 0.5 ~ 2.0 (default: 1.0)
• --output: Output MP3 file path

Available Voices

Voice ID	Language	Gender
female-tianmei	Chinese	Female
male-qn-qingse	Chinese	Male
male-qn-jianbin	Chinese	Male
English_expressive_narrator	English	Male

Returns: {"audio_path": "...", "size_bytes": ..., "duration_hint": "..."}

Environment

• MINIMAX_API_KEY — required
• MINIMAX_BASE_URL — optional (default: https://api.minimax.io)