QwenLM/Qwen3-ASR
Speech recognition and forced alignment for 52 languages
View on index · View in 3D Map
Related repos
- Transcribes audio into text across 52 languages
- Clones voices across languages with low latency
- open-source large language models for chat, code, and multilingual tasks
- Code-specific language model for coding tasks
- Bypasses API audio length limits via parallel processing
- Multimodal model with real-time speech synthesis
- Qwen2.5 language models in various sizes
- Accepts audio inputs for voice chat and analysis
- Speech recognition for 1600 languages
- MegaTTS3: speech-to-text aligner and TTS model with voice cloning, supports Chinese and English, runs locally
