qwen-audio-3.0-asr-flash

The Bailing large model ASR version updated in June 2026 fully supports the seven traditional Chinese dialect systems, including Mandarin, Wu, Xiang, Gan, Hakka, Min, and Cantonese, and adapts to over 20 regional Mandarin accents. It is specifically optimized for the rhythm, meter, and classical expression characteristics of ancient Chinese poetry to improve recognition accuracy, making it ideal for cultural heritage, educational narration, and audiobook scenarios. Punctuation prediction and text normalization capabilities are enhanced to align the output with written conventions, automatically converting numbers, dates, and amounts into standard formats to improve readability and professionalism. Language support has been expanded to 30 languages, including English, Japanese, Korean, Vietnamese, Thai, Indonesian, Malay, Filipino, Hindi, Arabic, French, German, Spanish, Portuguese, Russian, Italian, Dutch, Swedish, Danish, Finnish, Norwegian, Greek, Polish, Czech, Hungarian, Romanian, Bulgarian, Croatian, and Slovak. It also supports context capabilities and can transcribe audio up to 5 minutes in length.