11 families, 16 models

Voice

Voice models differ in how natural they sound, how much they act (whispers, laughter, pauses), how many languages they speak well, and price, from about half a credit to 25 credits per page read.

Our advice

Which one to take.

Or start from what you want to do

  • Read a note or narrate a filmKokoroThe default: private and by far the cheapest.Or ElevenLabs
  • A voice that acts: whispers, laughs, pausesGemini 3.1 Flash TTSSteer the delivery with tags written in the text.Or xAI TTS
  • Read in French without leaving private modeQwen3 TTSTen languages, French included, in private mode.Or Chatterbox HD

Quality and price

What you get for what you pay.

An Elo rating: higher sounds more natural to more people. 100 points apart means the higher-rated model wins about 64% of duels.

Artificial Analysis Speech Arena against priceElo rating · Artificial Analysis, October 16, 2026
Our advice Private Anonymized
Artificial Analysis Speech ArenaUS dollars per 1,000 characters (log scale)

Each point is a family’s best measured version. The top left holds the best value.

See the numbers
ModelArtificial Analysis Speech ArenaUS dollars per 1,000 characters (log scale)
ElevenLabs TTS v4 Turbo1,330 Elo$0.0207
Gemini 3.1 Flash TTS1,209 Elo$0.0843
Gradium TTS1,151 Elo$0.0214
MiniMax Speech-02 HD1,114 Elo$0.0562
Chatterbox HD (Resemble AI)1,102 Elo$0.0225
Kokoro Text to Speech1,064 Elo$0.00157

Leaderboards

Measured side by side.

Artificial Analysis Speech ArenaElo rating · Artificial Analysis, October 16, 2026
  1. 1. ElevenLabs TTS v4 TurboElevenLabs1,330
  2. 2. ElevenLabs TTS v4ElevenLabs1,329
  3. 3. Gemini 3.1 Flash TTSGemini 3.1 Flash TTS1,209
  4. 4. ElevenLabs TTS v3ElevenLabs1,178
  5. 5. Gradium TTSGradium TTS1,151
  6. 6. MiniMax Speech-02 HDMiniMax Speech-02 HD1,114
  7. 7. Chatterbox HD (Resemble AI)Chatterbox HD1,102
  8. 8. ElevenLabs Turbo v2.5ElevenLabs1,097
  9. 9. ElevenLabs Multilingual v2ElevenLabs1,095
  10. 10. Kokoro Text to SpeechKokoro1,064

People listen to the same sentence read by two voices, without knowing the models, and pick the more natural one.

See the numbers
ModelElo
1. ElevenLabs TTS v4 Turbo1,330
2. ElevenLabs TTS v41,329
3. Gemini 3.1 Flash TTS1,209
4. ElevenLabs TTS v31,178
5. Gradium TTS1,151
6. MiniMax Speech-02 HD1,114
7. Chatterbox HD (Resemble AI)1,102
8. ElevenLabs Turbo v2.51,097
9. ElevenLabs Multilingual v21,095
10. Kokoro Text to Speech1,064

Prices

Every family, side by side.

Every price, model by model (16)

One row per family, at its recommended version, in US dollars. Sort any column; tick up to three families to compare them in detail.

Nothing ticked yet.
Voice: every family at its recommended version
CompareElo ratingper 1,000 charactersPrivacy
ElevenLabsElevenLabs TTS v4 · ElevenLabs1,329$0.0414Anonymized
Gemini 3.1 Flash TTSGemini 3.1 Flash TTS · Google DeepMind1,209$0.0843Anonymized
Gradium TTSGradium TTS · Gradium1,151$0.0214Anonymized
MiniMax Speech-02 HDMiniMax Speech-02 HD · MiniMax1,114$0.0562Anonymized
Chatterbox HDChatterbox HD (Resemble AI) · Resemble AI1,102$0.0225Private
KokoroKokoro Text to Speech · hexgrad (open source)1,064$0.00157Private
Inworld TTS-1.5 MaxInworld TTS-1.5 Max · Inworld AINot measured$0.00562Anonymized
Orpheus TTSOrpheus TTS · Canopy Labs (open source)Not measured$0.0281Private
Qwen3 TTSQwen 3 TTS 1.7B · Alibaba (Qwen)Not measured$0.0506Private
xAI TTSxAI TTS v1 · xAINot measured$0.00843Anonymized
Seed Audio 1.0Seed Audio 1.0 · ByteDanceNot measured$0.0129 per 10 seconds of soundAnonymized

Release history

The latest versions.

  1. Eleven v4ElevenLabs
  2. Eleven v4 TurboElevenLabs
  3. Gradium TTSGradium TTS
  4. Gemini 3.1 Flash TTSGemini 3.1 Flash TTS
  5. xAI TTSxAI TTS
  6. Qwen3 TTSQwen3 TTS
  7. TTS-1.5 MaxInworld TTS-1.5 Max
  8. Eleven v3ElevenLabs

Prices read on October 16, 2026, in US dollars; scores dated on each chart. Sub Rosa shows the exact price before you start.