Introducing five new languages on the Voice Arena TTS leaderboard.
Models from @ElevenLabs rank #1 on all five of the new Voice Arena TTS leaderboards: French 🇫🇷, German 🇩🇪, Bahasa 🇮🇩, Tagalog 🇵🇭 and Indian English 🇮🇳.
Eleven v4 Turbo takes the top position on four of the five enLabs holds both #1 and #2, with Eleven v4 first and Eleven v4 Turbo second.
Across five languages spanning Europe, Southeast Asia and South Asia, each judged only by native listeners, ElevenLabs finishes first on every board, ahead of the next-best model from another provider by 11 to 26 Elo points.
Voice Arena ranks models through blind, head-to-head votes by vetted native speakers, scored with Bradley-Terry Elo.
Board-by-board results below 🧵
Models from @ElevenLabs rank #1 on all five of the new Voice Arena TTS leaderboards: French 🇫🇷, German 🇩🇪, Bahasa 🇮🇩, Tagalog 🇵🇭 and Indian English 🇮🇳.
Eleven v4 Turbo takes the top position on four of the five enLabs holds both #1 and #2, with Eleven v4 first and Eleven v4 Turbo second.
Across five languages spanning Europe, Southeast Asia and South Asia, each judged only by native listeners, ElevenLabs finishes first on every board, ahead of the next-best model from another provider by 11 to 26 Elo points.
Voice Arena ranks models through blind, head-to-head votes by vetted native speakers, scored with Bradley-Terry Elo.
Board-by-board results below 🧵
39
19
138
409,540
🇮🇩 Bahasa
ElevenLabs takes both of the top two places: Eleven v4 at #1 of 6 with 1118 Elo, and Eleven v4 Turbo at #2 with 1107.
voicearena.com/tts-leaderboar…
ElevenLabs takes both of the top two places: Eleven v4 at #1 of 6 with 1118 Elo, and Eleven v4 Turbo at #2 with 1107.
voicearena.com/tts-leaderboar…
1
1
4
1,185
Explore the full leaderboards on Voice Arena → voicearena.com/tts-leaderboard
0
0
3
658
🚀 Mistral AI’s Voxtral Mini Transcribe 3 ranks #2 out of 9 models by DER on Voice Arena’s Diarization + ASR leaderboard, outperforming its larger counterpart.
Following the launch of Mistral’s new Voxtral 3 family, we evaluated its Mini and Small transcription models on Voice ch.
🥈 Voxtral Mini Transcribe 3: 40.4% DER (#2)
Voxtral Small Transcribe 3: 41.1% DER (#4)
Both models were tested on 139 held-out conversations under identical conditions, using human-verified references, a strict 0 ms collar, and no predefined speaker counts or audio cleanup.
Mini leads on in-person conversations (40.6% vs. 41.5%), while Small performs slightly better on online calls (39.1% vs. 39.5%).
Overlapping speech remains the biggest challenge. Missed speech accounts for 82% of Mini's total DER. Both models correctly identify the exact number of speakers in 83.5% of sessions.
Congratulations to @MistralAI on the release ! 👏
View complete track & condition breakdown: voicearena.com/diarization-be…
Following the launch of Mistral’s new Voxtral 3 family, we evaluated its Mini and Small transcription models on Voice ch.
🥈 Voxtral Mini Transcribe 3: 40.4% DER (#2)
Voxtral Small Transcribe 3: 41.1% DER (#4)
Both models were tested on 139 held-out conversations under identical conditions, using human-verified references, a strict 0 ms collar, and no predefined speaker counts or audio cleanup.
Mini leads on in-person conversations (40.6% vs. 41.5%), while Small performs slightly better on online calls (39.1% vs. 39.5%).
Overlapping speech remains the biggest challenge. Missed speech accounts for 82% of Mini's total DER. Both models correctly identify the exact number of speakers in 83.5% of sessions.
Congratulations to @MistralAI on the release ! 👏
View complete track & condition breakdown: voicearena.com/diarization-be…
11
3
45
4,335
Murf AI's Falcon 2 moves from #11 to #2 of 14 on the Voice Arena US English 🇺🇸 latency board, with 102 ms to first audio at the median.
That is more than five times faster than its own earlier 528 ms, and puts Falcon 2 ahead of SpeechifyAI's Simba 3.2 at 123 ms and Inworld TTS 2 e time to first audio the way a listener hears it: leading silence counts, and no exact text is ever sent twice.
We're also evaluating Murf's latest Falcon 2 release in US English and Hindi, with results coming soon.
Congratulations to @MurfAIStudio on the update!
Explore the board → voicearena.com/tts-leaderboar…
That is more than five times faster than its own earlier 528 ms, and puts Falcon 2 ahead of SpeechifyAI's Simba 3.2 at 123 ms and Inworld TTS 2 e time to first audio the way a listener hears it: leading silence counts, and no exact text is ever sent twice.
We're also evaluating Murf's latest Falcon 2 release in US English and Hindi, with results coming soon.
Congratulations to @MurfAIStudio on the update!
Explore the board → voicearena.com/tts-leaderboar…
4
6
27
1,590
Mistral AI's Voxtral-mini-tts-3-0 debuts at #10 of 23 on the Voice Arena US English 🇺🇸 TTS Leaderboard at 1011 Elo, ahead of Gradium-TTS, Fish Audio's Drama 3 and xAI's Grok TTS.
Voxtral-mini-tts-3-0 is the latest TTS model from @MistralAI, part of the Voxtral 3 release, with es and 50+ dialects. Every ranking on Voice Arena comes from blind, head-to-head votes by vetted native speakers, scored with Bradley-Terry Elo.
It also ranks #6 of 20 on our Hindi board, making it one of seven models in the top 10 on both boards.
Results backed by 700+ blind, head-to-head listener votes from native US English speakers.
Congratulations to @MistralAI on the release!
See below for the exact audio raters judged and unedited listener verdicts 🧵
Voxtral-mini-tts-3-0 is the latest TTS model from @MistralAI, part of the Voxtral 3 release, with es and 50+ dialects. Every ranking on Voice Arena comes from blind, head-to-head votes by vetted native speakers, scored with Bradley-Terry Elo.
It also ranks #6 of 20 on our Hindi board, making it one of seven models in the top 10 on both boards.
Results backed by 700+ blind, head-to-head listener votes from native US English speakers.
Congratulations to @MistralAI on the release!
See below for the exact audio raters judged and unedited listener verdicts 🧵
2
9
32
44,201
And the male voice, on a public service announcement.
A benefits notice for Cuyahoga County: a hard-to-pronounce place name, two acronyms and a batch code. SNAP is spoken as a word and EBT letter by letter, and a voice has to know which is which 🔊
A benefits notice for Cuyahoga County: a hard-to-pronounce place name, two acronyms and a batch code. SNAP is spoken as a word and EBT letter by letter, and a voice has to know which is which 🔊
1
0
5
130
What the raters typed after voting, unedited 👇
No self-reported benchmarks anywhere in this. Blind pairs, vetted native ears, Bradley-Terry math. That's the whole system.
Congrats to @arthurmensch and the whole @MistralAI team 👏
Explore the board → voicearena.com/tts-leaderboar…
No self-reported benchmarks anywhere in this. Blind pairs, vetted native ears, Bradley-Terry math. That's the whole system.
Congrats to @arthurmensch and the whole @MistralAI team 👏
Explore the board → voicearena.com/tts-leaderboar…
0
0
5
78
Mistral AI's Voxtral-mini-tts-3-0 debuts at #6 of 20 on the Voice Arena Hindi 🇮🇳 TTS Leaderboard at 1048 Elo, ahead of Maya Research's Maya-2-Native, ElevenLabs' Eleven v3 and Cartesia's Sonic-3.5.
Voxtral-mini-tts-3-0 is the latest TTS model from @MistralAI , part of the support for 35+ languages and 50+ dialects. Every ranking on Voice Arena comes from blind, head-to-head votes by vetted native speakers, scored with Bradley-Terry Elo.
Only Google DeepMind, ElevenLabs and Cartesia place models above it, which makes it the highest-ranked model on the Hindi board from any other lab, in a tight group where Maya-2-Native and Eleven v3 sit within 4 points.
Results backed by 600+ blind, head-to-head listener votes from native Hindi speakers.
Voxtral-mini-tts-3-0 is also being evaluated on our US English board, with results coming soon.
Congratulations to @MistralAI on the release!
See below for the exact audio raters judged and unedited listener verdicts 🧵
Voxtral-mini-tts-3-0 is the latest TTS model from @MistralAI , part of the support for 35+ languages and 50+ dialects. Every ranking on Voice Arena comes from blind, head-to-head votes by vetted native speakers, scored with Bradley-Terry Elo.
Only Google DeepMind, ElevenLabs and Cartesia place models above it, which makes it the highest-ranked model on the Hindi board from any other lab, in a tight group where Maya-2-Native and Eleven v3 sit within 4 points.
Results backed by 600+ blind, head-to-head listener votes from native Hindi speakers.
Voxtral-mini-tts-3-0 is also being evaluated on our US English board, with results coming soon.
Congratulations to @MistralAI on the release!
See below for the exact audio raters judged and unedited listener verdicts 🧵
2
8
23
53,076
And the male voice, on a celebrity story.
A star sighting outside a designer store, told like gossip between friends: brand and place names, "like" and "literally", in mixed Hindi and English. Fast, casual storytelling tests whether a voice keeps its energy without rushing 🔊
A star sighting outside a designer store, told like gossip between friends: brand and place names, "like" and "literally", in mixed Hindi and English. Fast, casual storytelling tests whether a voice keeps its energy without rushing 🔊
1
0
5
144
What the raters typed after voting, unedited, including a matchup where it beat Google DeepMind's Gemini 3.1 Flash TTS 👇
No self-reported benchmarks anywhere in this. Blind pairs, vetted native ears, Bradley-Terry math. That's the whole system.
Congrats to @arthurmensch and am 👏
Explore the board → voicearena.com/tts-leaderboar…
No self-reported benchmarks anywhere in this. Blind pairs, vetted native ears, Bradley-Terry math. That's the whole system.
Congrats to @arthurmensch and am 👏
Explore the board → voicearena.com/tts-leaderboar…
0
0
3
105
Congratulations to @GradiumAI on the open-source release of Phonon, a family of text-to-speech models under 50M parameters. The smallest model, Phonon-31m, is 64 MB and is released under Apache 2.0.
Open releases like this give the whole field a stronger foundation to build on.
Gradium's cloud model, Gradium-TTS, already records the lowest median time to first audio among the cloud TTS APIs we measure, at 68 ms. Phonon begins speaking in 7.5 ms on a Snapdragon 8 Elite Gen 5, with no network hop involved.
On-device inference also changes the operating constraints. Once the model and a voice are downloaded, no connection is required, the text never leaves the device, and usage is unlimited at no extra cost. Launch partners are already applying it to game characters and automotive R&D.
The family covers five languages and is supported at launch in Hugging Face Transformers, MLX-Audio and Ollama. Phonon-31m runs faster than real time on a Raspberry Pi 5.
A significant contribution to open, on-device speech.
Open releases like this give the whole field a stronger foundation to build on.
Gradium's cloud model, Gradium-TTS, already records the lowest median time to first audio among the cloud TTS APIs we measure, at 68 ms. Phonon begins speaking in 7.5 ms on a Snapdragon 8 Elite Gen 5, with no network hop involved.
On-device inference also changes the operating constraints. Once the model and a voice are downloaded, no connection is required, the text never leaves the device, and usage is unlimited at no extra cost. Launch partners are already applying it to game characters and automotive R&D.
The family covers five languages and is supported at launch in Hugging Face Transformers, MLX-Audio and Ollama. Phonon-31m runs faster than real time on a Raspberry Pi 5.
A significant contribution to open, on-device speech.
A new frontier for on-device speech.
We're open-sourcing Phonon, a suite of TTS models under 50m parameters, setting a new standard for naturalness and accuracy across languages, with the broadest inference support to date.
Phonon-31m is available under Apache 2.0.
Y🧵
6
7
54
120,054
Insights from training on Monsoon: More data continues to improve performance, even at 3,500 hours.
We trained the same model using the same recipe on randomly sampled 1,000, 2,000, and ~3,500-hour subsets of Monsoon across Hindi, Tamil, and Telugu.
Increasing training data rs reduced errors by:
• 11–14% on IndicVoices (unseen conversations)
• 21–25% on Vistaar
The gains diminish with each additional hour of data, but none of the six learning curves has flattened yet.
The takeaway: We haven’t hit the data ceiling for these languages. More high-quality conversational data continues to deliver meaningful improvements.
Monsoon is Voice Arena’s ASR training corpus, built to help bring word error rates down to single digits across 1,000 languages. It is available for licensing today in 50 languages.
See detailed results here: voicearena.com/datasets#resul…
We trained the same model using the same recipe on randomly sampled 1,000, 2,000, and ~3,500-hour subsets of Monsoon across Hindi, Tamil, and Telugu.
Increasing training data rs reduced errors by:
• 11–14% on IndicVoices (unseen conversations)
• 21–25% on Vistaar
The gains diminish with each additional hour of data, but none of the six learning curves has flattened yet.
The takeaway: We haven’t hit the data ceiling for these languages. More high-quality conversational data continues to deliver meaningful improvements.
Monsoon is Voice Arena’s ASR training corpus, built to help bring word error rates down to single digits across 1,000 languages. It is available for licensing today in 50 languages.
See detailed results here: voicearena.com/datasets#resul…
3
1
8
945
Chariot TTS from Chariot is now on the @voicearena_ai Hindi 🇮🇳 TTS Leaderboard, entering at #12 of 19 with 972 Elo.
Every ranking on Voice Arena comes from blind, head-to-head votes by vetted native speakers, scored with Bradley-Terry Elo.
It enters with a displayed rank range statistical band of Mirai Minds' Mira-TTS and Cartesia's Sonic-3. Its strongest category is Media & Entertainment, where it sits at #11 of 19.
Results backed by 1,300+ blind, head-to-head listener votes from native Hindi speakers.
Congratulations to the Chariot team!
Explore the board → voicearena.com/tts-leaderboar…
Every ranking on Voice Arena comes from blind, head-to-head votes by vetted native speakers, scored with Bradley-Terry Elo.
It enters with a displayed rank range statistical band of Mirai Minds' Mira-TTS and Cartesia's Sonic-3. Its strongest category is Media & Entertainment, where it sits at #11 of 19.
Results backed by 1,300+ blind, head-to-head listener votes from native Hindi speakers.
Congratulations to the Chariot team!
Explore the board → voicearena.com/tts-leaderboar…
1
1
5
475