Announcing Multilingual Text to Speech Arena Leaderboards, comparing leading TTS models across 9 languages beyond English 🇯🇵 🇨🇳 🇮🇳 🇪🇸 🇩🇪 🇫🇷 🇵🇹 🇻🇳 🇸🇦
Text to Speech performance varies significantly across languages, with models that perform well in English not necessarily delivering the same pronunciation, pacing, tone, and naturalness in other languages.
We have extended the Artificial Analysis Controlled Voice Arena to 9 new languages: Japanese 🇯🇵, Mandarin Chinese 🇨🇳, Hindi 🇮🇳, Spanish 🇪🇸, German 🇩🇪, French 🇫🇷, Portuguese 🇵🇹, Vietnamese 🇻🇳, and Arabic 🇸🇦.
Each model is evaluated using standardized cloned voices, with prompts written natively in each language and preference votes collected from first-language speakers. Elo scores are calculated independently for each language, allowing each leaderboard to reflect model preferences among speakers of that language.
Key results:
➤ @Cartesia's Sonic family leads 8 of the 9 new language leaderboards, with Sonic 3.6 ranking #1 in 7 languages and Sonic 3.5 leading Portuguese. @InworldAI's Realtime TTS-2 takes the #1 spot in Mandarin.
➤ @ElevenLabs' Eleven v3 family ranks in the top 3 across 7 of…
Artificial Analysis extended its Controlled Voice Arena to 9 new languages (Japanese, Mandarin, Hindi, Spanish, German, French, Portuguese, Vietnamese, Arabic), scored with natively written prompts and votes from first-language speakers.
Text-to-speech quality doesn't transfer reliably across languages: models that perform well in English don't necessarily match it on pronunciation, pacing, tone, and naturalness elsewhere.
Cartesia's Sonic family leads 8 of the 9 new leaderboards (Sonic 3.6 ranks #1 in 7 languages, Sonic 3.5 leads Portuguese), while Inworld's Realtime TTS-2 takes #1 in Mandarin at 1,185 Elo, ahead of Sonic 3.6 (1,146) and StepFun's StepAudio 2.5 TTS (1,130).
ElevenLabs' Eleven v3 family (Eleven v3 and Eleven v3 Conversational) places in the top 3 across 7 of the 9 new languages, showing consistency across languages rather than one strong market.
The rankings draw on more than 100,000 human preference votes across the 9 languages, with 15 to 24 public models ranked per language.
Read the full post on X ↗
