New TTS benchmark tests 26 models across 30,564 requests: fastest at 46ms
mattturck · x · 2026-10-20
Cekura's Text-to-Speech Bench fired 30,564 requests at 26 TTS models, comparing speed, reading accuracy, naturalness, and interruption handling.
- Median time to first audio for a full sentence: 46ms for the fastest model vs 1.65s for the slowest — a 36x gap
- Fastest ≠ most natural: the top-naturalness model ranked only 15th of 26 on full-sentence speed
- Methodology: same 104 agent sentences read 3x per model from one cloud region, plus word-by-word streaming, interruption handling, playback gaps, and failure tests
- Gradium TTS leads streaming speed (64ms, 4.51 naturalness); 15 of 17 streaming-capable models are both slower and less natural
A practical selection reference for voice-agent builders.
More from Models
- Deutsche Bank: Meta's Muse still leads ChatGPT in downloads and DAUs 36 days post-launch — alexandr_wang · 2026-10-20
- Uncensored DeepSeek V4.1 Flash FP8 finetune trends on Hugging Face — dealignai · 2026-10-20
- ChatGPT caught silently swapping GPT-5.6 High for GPT-6 at medium thinking — DepositameAMiBCP · 2026-10-20
- MiniMax M3.1 Flash goes live in Command Code with 1M context, free for now — MiniMax_AI · 2026-10-20
- User slams Xiaomi MiMo over hard-to-cancel $60 subscription — bookwormengr · 2026-10-20
- Strata 2 beta runs at half speed? Dual-4090 user reports 155→95 tok/s — Lost-Vermicelli-6252 · 2026-10-20