GO
| HSI1 | 25,087.75 | +45.04 | 249.21B |
| HSCEI1 | 8,362.60 | +22.97 | 88.82B |
| Back Zoom + Zoom - Block Traded | |
|
BABA-W Upgrades Qwen-Audio-3.1 Voice Model Family
2026-09-22 12:24:04 BABA-W (09988.HK) announced at the Apsara Conference a full-fledged upgrade to its Qwen-Audio-3.1 voice model family, covering three major series: automatic speech recognition (ASR), text-to-speech (TTS) and real-time voice interaction (Realtime). The next-generation audio understanding model, Qwen-Audio-3.1-ASR-Next, was newly launched. The model can understand human emotions, music, environmental sounds and mechanical noises, enabling sound description, event localization, audio question answering and reasoning, allowing it to accurately answer questions such as the emotional state of people in a recording. The audio creation model Qwen-Audio-3.1-TTS-Next also debuted with a new architecture. Based on a script, it can directly generate complete audio integrating dialogue and ambient sounds, considerably improving production efficiency for audiobooks, film and television productions, and podcasts. ~ AASTOCKS Financial News Website: www.aastocks.com | |