Hugging Face releases Open TTS Leaderboard
Understand the story
On September 30, 2026, Hugging Face announced in an official blog post (primary source) the launch of the Open TTS Leaderboard. The leaderboard evaluates open-source TTS (text-to-speech) models using an objective metric system: WER/CER computed with Qwen3 ASR measures speech intelligibility, RTFx and TTFA measured on H200 reflect inference speed and latency, and WavLM is used to assess speaker similarity, covering multilingual TTS and voice cloning scenarios. According to the report, this objective evaluation pipeline cuts the evaluation time per model from about two weeks to a few hours, significantly improving efficiency. The event is currently at the official Hugging Face launch stage, with no further developments reported.
AI overview of reporting · Updated 17 days ago
Reporting timeline
Explore different aspects of the story through its reporting.
- Hugging Face BlogSelectedOpen TTS Leaderboard: Scalable Evaluation for Multilingual Text-to-Speech and Voice Cloning
Hugging Face has launched the Open TTS Leaderboard, evaluating open-source TTS models with objective metrics including Qwen3 ASR WER/CER, RTFx and TTFA on H200, and WavLM speaker similarity, cutting per-model evaluation time from about two weeks to hours.