Open TTS Leaderboard: Scalable Evaluation for Multilingual Text-to-Speech and Voice Cloning
Comparing speech synthesis systems across languages and voices has lacked standardized metrics, making it harder to track progress in this rapidly advancing field.
Get smart on it
The Open TTS Leaderboard is a tool for evaluating text-to-speech models using objective metrics like word error rate, processing speed, and speaker similarity, rather than relying solely on human preference voting. Existing arena-based leaderboards that compare models through user votes cannot keep pace with the rapid release of open-source TTS models and face challenges with voter consistency over time. This leaderboard addresses those limitations by using automated metrics that reduce evaluation time from weeks to hours, though the creators acknowledge these metrics do not directly measure naturalness or listener preference. The leaderboard supports multilingual evaluation and voice cloning comparisons, and includes a separate section where users can listen to and vote on model outputs to complement the objective metrics.
Related news
With most information hidden, the game Stratego had stumped AI, until now
Deep Blue took down Garry Kasparov at chess in 1997, AlphaGo beat Lee Sedol at Go in 2016, and poker bots have been beating professionals for years. But one classic game called Stratego held out. Even DeepMind, with its exceptional budget, couldn't build a machine that reliably beat the best human players. Now, a team of researchers from Carnegie Mellon, MIT, New York University, and Stanford University has done it. Their AI, called Ataraxos, beat Pim Niemeijer, arguably the best Stratego player
Opus 5.5 loves to tell you ‘this matters’ (and other AI writing tells)
Opus 5.5’s biggest tell is the word “dependable,” which pops up 23 times more often than in human samples.
