Benchmark › Speech models › Eleven v3
ElevenLabs · tested 2026-11-22
Eleven v3
82.3overall · rank 9 of 19 speech models
API model: eleven_v3
Reached through Replicate
Voice: Sarah44.1 kHz
Profile
Each spoke is one test, 0 at the centre and 100 at the edge.
Eleven v3Median of all models (dashed)
Test by test
The dot shows where this model sits among all 19: left is last, right is first.
Long text
One document sent at five lengths, a single request each. Longest clean read: 4,000 characters.
| 1,000 characters | 2,000 characters | 4,000 characters | 8,000 characters | 15,000 characters |
|---|---|---|---|---|
| 100 | 100 | 100 | rejected | rejected |
Numbers, dates and symbols
60 of 60 sentences read correctly. Hover a square for the sentence.
Emotion control
9 of 40 directed clips measurably changed. Control used: inline tag.
Angry
0/10
Excited
3/10
Scared
6/10
Whisper
0/10
Other languages
Audio from the test
Unedited clips exactly as the model returned them.
Plain sentence
TextThe ferry leaves the harbor every morning just after sunrise, and most of the passengers are still half asleep.
Repetition
TextNo, no, no, no, no. I said no.
Money amount
TextThe total comes to $1,250.50 after tax.
Spanish
Japanese
Neutral reading
Asked for: angry
Asked for: excited
Asked for: sad (not scored yet)
Asked for: whisper
Compare with
Gemini 3.8 Flash-Lite TTS · 99.2AnyToSpeech Standard voices · 98.9Gemini 3.8 Flash TTS · 98.1AnyToSpeech Premium and Characters · 93.6MiniMax Speech 2.8 HD · 90.3MiniMax Speech 2.8 Turbo · 89.2Grok TTS · 88.1Inworld TTS-2 · 84.8GPT-4o mini TTS · 81.6Kokoro 82M · 81.6Eleven Flash v2.5 · 81.3Eleven Multilingual v2 · 79Inworld TTS-1.5 Max · 77.1Gemini 2.5 Pro TTS · 76.8Qwen3 TTS · 76.3TTS-1 · 76.2TTS-1 HD · 70.9Chatterbox Multilingual · 59.9
Think a score is wrong? Check it.
Every test item, scoring rule and known limit is published, and the audio behind each number is on the page.