Alibaba Drops Pricing for Qwen Audio’s 3.1 Voice Assistants
New Qwen-Audio-3.1 speech recognition and synthesis models were published by Alibaba, alongside a price drop (TTS — ~70%, real-time — ~85%, ASR — up to 95%).
Alibaba released 5 new audio models from its Qwen-Audio-3.1 series: ASR and ASR-Next for multi-lingual/dialect speech recognition, speaker/emotion identification; TTS and TTS-Next for text-to-speech synthesis and cross-language voice conversion with controllable tone via plain text. Meanwhile, they also cut down the cost (TTS ~70%, realtime ~85%, ASR up to 95%).
These models are cheap but not powerful (just a feature list and pricing for the new APIs without any benchmark results). So I’m still interested in these kinds of news because the voice UI part is usually the most expensive part in building agents, and a drop of 70–95% on various models will save us lots of money. But remember that all these are from Alibaba, you need to validate them before using them.
Source: the-decoder.com
Free course
Stop reading about AI — start building with it
The free Claude Code course: your first site, tool or game — no coding. No upsells, no cross-sells — nothing to buy here.
Start free →▌ Related guides
Author
Evgenii Arsentev
PhD · Founder & CEO, Dad4Kids
Articles · Latest articles