Alibaba’s Qwen team released Qwen-Audio-3.1, a five-model voice AI stack covering automatic speech recognition (ASR), text-to-speech (TTS), and real-time conversational interaction, alongside two new models: TTS-Next for audio creation with emotional control and sound effects, and ASR-Next for audio understanding. The ASR upgrade improves multilingual and dialect recognition while automatically stripping filler words. Alibaba simultaneously cut prices across the lineup, with ASR down as much as 95%, TTS down about 70%, and Realtime down roughly 85%.
