AIQwen-Audio-3.1 announced with five models and price cuts on September 23, 2026
Qwen says the update adds TTS-Next for audio creation and ASR-Next for audio understanding, alongside upgrades to speech recognition, speech synthesis and real-time interaction.
TLDR
Qwen’s September 23, 2026 announcement introduced a five-model audio lineup. The company says TTS-Next generates voices, sound effects and background audio in one pass, while ASR-Next supports multi-speaker transcription with speaker labels and timestamps, plus understanding of emotions and environmental sounds.
Qwen also says its upgraded speech recognition can remove fillers and repetitions, its speech synthesis accepts instructions for emotion, speed and style, and Realtime supports simultaneous speaking and listening with interruptions at any time.
The announcement advertised price cuts of about 70% for TTS, about 85% for Realtime and up to 95% for ASR. Qwen said more APIs were coming soon.
