Try AnyToSpeech
Benchmark › Speech models › Long text
15% of the overall score · updated 2026-11-22

Long text

How much text can it take in one request and still read every word?

Word accuracyNumbers, dates and symbolsLong textRepeated wordsOther languagesEmotion controlReliabilitySpeed

Ranking

Score out of 100, higher is better. 5 items per model; the thin line on each bar is the 95% range.

0255075100

When two models' ranges overlap, this test cannot tell them apart.

Longest clean read

The largest single request each model read with at least 97% of words correct. Longer is better.

07.5k characters15k

Length by length

Word accuracy at each request size; darker is better. "Rejected" means the model refused that much text in one request. Hover a cell for details.

ModelScore1,000 chars2,000 chars4,000 chars8,000 chars15,000 chars
1AnyToSpeech Premium and Charactersclean up to 15,000 100 100100100100100
2Kokoro 82Mclean up to 15,000 100 100100100100100
3Eleven Flash v2.5clean up to 15,000 100 10010010099.9100
4Gemini 3.8 Flash-Lite TTSclean up to 15,000 99.9 10010099.999.999.8
5AnyToSpeech Standard voicesclean up to 15,000 99.9 10010099.699.999.8
6Gemini 3.8 Flash TTSclean up to 15,000 99.9 10099.510099.999.9
7Grok TTSclean up to 15,000 99.8 10099.799.699.899.9
8MiniMax Speech 2.8 HDclean up to 8,000 79.9 10010099.7100rejected
9MiniMax Speech 2.8 Turboclean up to 8,000 79.9 10099.799.799.9rejected
10Eleven Multilingual v2clean up to 8,000 79.9 10010099.799.7rejected
11GPT-4o mini TTSclean up to 8,000 79.8 10099.799.699.7rejected
12Eleven v3clean up to 4,000 60 100100100rejectedrejected
13TTS-1 HDclean up to 4,000 60 100100100rejectedrejected
14Gemini 2.5 Pro TTSclean up to 4,000 59.9 10010099.7rejectedrejected
15TTS-1clean up to 4,000 59.9 10010099.6rejectedrejected
16Qwen3 TTSclean up to 2,000 58.6 10099.562.129.51.9
17Inworld TTS-2clean up to 2,000 40 100100rejectedrejectedrejected
18Inworld TTS-1.5 Maxclean up to 2,000 40 100100rejectedrejectedrejected
19Chatterbox Multilingualclean up to nothing 13.3 34.317.28.64.12.1
50 or lower100

Products built on these models, AnyToSpeech included, split long documents into pieces before sending them. This test shows what the model itself does with one long request.

What we test and how it is scored

Test. One 15,000-character document, sent at five lengths: about 1,000, 2,000, 4,000, 8,000 and 15,000 characters. Each length is a single request with no splitting on our side, so every model is pushed up to and past its own limit.

Scoring. Each recording is transcribed in one-minute pieces and compared word for word with the text sent. A length scores its word accuracy; a length the model rejects, cuts short or garbles scores what comes back, which is 0 when nothing usable does. Score = average across the five lengths. "Longest clean read" is the largest length read with at least 97% of words correct.

Full item list and rules: methodology.

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.

Read the methodology