Omni SenseVoice
lastmod 2024-11-11created 2024-11-111 edits
Omni SenseVoice: High-Speed Speech Recognition with words timestamps 🗣️🎯
High-speed speech recognition with per-word timestamps
Features
- Based on SenseVoice, a multilingual speech understanding model with automatic speech recognition (ASR), spoken language identification (LID), speech emotion recognition (SER), and audio event detection (AED)
- Optimized for ultra-fast inference and accurate timestamps, enabling smarter and faster audio transcription
- Key options
-
--language- auto-detect/specify the language (auto, zh, en, yue, ja, ko) -
--textnorm- choose whether to apply inverse text normalization (withitn for inverse-normalized, woitn for raw text) -
--device-id- run on a specific GPU (default: -1 for CPU) -
--quantize- use a quantized model for faster processing
-