0-ASR to create a new model family.
Audio-language model suite adds speaker-aware transcription and understanding
Qwen-Audio-3.1-ASR builds on its predecessor with improved multilingual ASR, structured speaker-diarized outputs, and general audio understanding capabilities.
Chinese Tech
Chuanmeng Bian · Daren Chen · Peixin Chen · Zhigao Chen · Chong Deng · Zhiyun Fan · +42 more
Alibaba Group
Research Digest··3 min read
1-ASR, a family of audio-language models that extend accurate speech recognition toward structured transcription and broader audio interpretation.
§