Beyond the Benchmark: Conditioning, Domains and Failure Modes of Qwen3-TTS
Deep technical analysis of TTS model behavior for researchers working on speech synthesis and voice cloning systems.
AI Summary
Researchers analyze Qwen3-TTS voice cloning performance across domains, finding neither conditioning method dominates and revealing specific failure modes in air traffic control speech.
Excerpt
Qwen3-TTS reports state-of-the-art zero-shot voice cloning, but its technical report leaves open how the model should be conditioned, how it behaves outside read speech, and how it fails. We characterise its two cloning interfaces, in-context learning (ICL) and X-Vector, against Chatterbox, XTTS-v2 and F5-TTS on read, meeting, emotional and air-traffic-control (ATC) speech, using eighteen objective measures, an LLM judge, a spoofing countermeasure and a listening test. Neither interface dominate