Open-source NAR TTS scores WER 1.30 (EN) / 0.99 (ZH) on Seed-TTS, seeks cursed test cases

Important_Drag_6890 · reddit · 2026-09-24

The ViiTorVoice team reports benchmark results for their open-source NAR TTS model on Seed-TTS test sets: 1.30 English WER, 0.99 Chinese WER, and 0.734 English speaker similarity. Model and inference code are on GitHub. Noting that benchmarks cover only a clean slice of real usage, they are collecting harder cases — mixed Chinese-English text, unusual names, numbers/acronyms/CLI commands, tongue twisters, long-form narration, and weird punctuation — and inviting users to submit sentences that break TTS systems.

Original post →

More from Models

Models channel →