{"version":1,"type":"story","url":"https://digestai.news/story/hugging-face-launches-open-tts-leaderboard-for-multilingual-text-to-sp","json":"https://digestai.news/story/hugging-face-launches-open-tts-leaderboard-for-multilingual-text-to-sp.json","markdown":"https://digestai.news/story/hugging-face-launches-open-tts-leaderboard-for-multilingual-text-to-sp.md","slug":"hugging-face-launches-open-tts-leaderboard-for-multilingual-text-to-sp","headline":"Hugging Face launches Open TTS Leaderboard for multilingual text-to-speech evaluation","summary":"Hugging Face introduced the Open TTS Leaderboard on September 30, 2026, to address gaps in text-to-speech (TTS) model evaluation. Existing arena-style leaderboards rely on human preference scores but struggle with scalability and open-source representation, listing only 16 of 92 models as open-weights. The new leaderboard uses objective metrics—word/character error rates (WER/CER), inference speed (RTFx, TTFA), and speaker similarity (SIM)—to rank models on intelligibility, performance, and voice cloning. It covers English, Chinese, Japanese, and Korean, with Pareto plots showing trade-offs between metrics.\n\nThe leaderboard prioritizes open-source models and multilingual support, offering a 'Listen' tab for direct audio comparisons and a 'Streaming' tab for latency benchmarks. Models like `hexgrad/Kokoro-82M` and `fishaudio/s2-pro` lead in English WER, while `k2-fsa/OmniVoice` excels in multilingual performance. Hugging Face plans to open-source evaluation scripts and invites community feedback to refine future iterations.","keyPoints":["Open TTS Leaderboard uses objective metrics (WER, RTFx, SIM) instead of human preference for faster, scalable evaluation","Only 16 of 92 TTS models on arena-style leaderboards are open-weights, per September 30 data","Models like `hexgrad/Kokoro-82M` rank top in English WER, but multilingual performance varies by language"],"whyItMatters":"The leaderboard accelerates TTS model development by providing standardized, objective benchmarks for open-source and multilingual models, addressing gaps in current evaluation methods.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":[],"models":["hexgrad/Kokoro-82M","Supertone/supertonic-3","fishaudio/s2-pro","k2-fsa/OmniVoice","FunAudioLLM/Fun-CosyVoice3-0.5B-2512","bosonai/higgs-tts-3-4b"],"people":[]},"firstPublishedAt":"2026-09-30T00:00:00Z","updatedAt":"2026-09-30T00:00:00Z","sourceCount":1,"hasPrimarySource":true,"sources":[{"outlet":"Hugging Face","title":"Open TTS Leaderboard: Scalable Evaluation for Multilingual Text-to-Speech and Voice Cloning","url":"https://huggingface.co/blog/open-tts-leaderboard","publishedAt":"2026-09-30T00:00:00Z","type":"primary","primary":true,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Hugging Face launches Open TTS Leaderboard for multilingual text-to-speech evaluation\", 30 September 2026, https://digestai.news/story/hugging-face-launches-open-tts-leaderboard-for-multilingual-text-to-sp","publisher":"Digest AI","title":"Hugging Face launches Open TTS Leaderboard for multilingual text-to-speech evaluation","datePublished":"2026-09-30T00:00:00Z","url":"https://digestai.news/story/hugging-face-launches-open-tts-leaderboard-for-multilingual-text-to-sp"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}