analiza aiqbench
This study provides a rigorous empirical evaluation of Qwen3-TTS, revealing that domain characteristics significantly outweigh system choice in determining performance variance across diverse speech types. While the model achieves state-of-the-art zero-shot cloning for read and meeting scenarios, it exhibits critical failure modes in air-traffic-control environments and struggles with prosody compared to its intelligibility scores. The findings offer essential insights into the practical limitations of current voice cloning interfaces regarding robustness against transcription errors and spoofing detection.
Beyond the Benchmark: Conditioning, Domains and Failure Modes of Qwen3-TTS
arXiv:2610.23678v1 Announce Type: cross Abstract: 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 dominates: ICL clones timbre more closely from correctly transcribed references, X-Vector clones timing, and ICL degrades sharply when a short reference transcript is wrong. Domain explains four times more variance than system choice. Qwen3-TTS leads on intelligibility and pitch register but trails on prosody, and the objective measures rank systems differently from listeners. An off-the-shelf countermeasure flags its clones no more often than genuine recordings, and on ATC radio it fails for every system. Code and audio are released.
Abstract: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 dominates: ICL clones timbre more closely from correctly transcribed references, X-Vector clones timing, and ICL degrades sharply when a short reference transcript is wrong. Domain explains four times more variance than system choice. Qwen3-TTS leads on intelligibility and pitch register but trails on prosody, and the objective measures rank systems differently from listeners. An off-the-shelf countermeasure flags its clones no more often than genuine recordings, and on ATC radio it fails for every system. Code and audio are released.
Sound (cs.SD); Computation and Language (cs.CL)
arXiv:2610.23678 [cs.SD]
(or arXiv:2610.23678v1 [cs.SD] for this version)
https://doi.org/10.48550/arXiv.2610.23678 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 244 cuvinte
Subiecte conexe
Projection sampling preserves capabilities but diminishes gains from alignment fine-tuning
LessWrong · 16 oct. 2026
Closing the Binary Blind Spot: Domain-Knowledge Injection for LLM Malware Triage in npm Packages
Databricks · 16 oct. 2026
Direct Preference Optimization Beyond Chatbots
Hugging Face · 3 iun. 2026
[AINews] High Taste Testers Ep. 1
Latent Space · 17 oct. 2026