WoodstitchHuman evals for voice AI

TTS voice leaderboard · US English · preliminary

Best TTS Voices for Nonfiction Narration

Stock text-to-speech voices ranked by blind paired listening on nonfiction book passages.

Voices53 from 19 providersJudgments550 so farListeners42 paid, USUpdated20 Nov 2026, 20:30 UTCHow we evaluate →

Top 20 voices

Elo-scale strength from blind paired listening (1000 = average voice in this use case) · whiskers = 95% interval · higher is better

#RangeVoiceScoreEloJudgmentsPlay
11–28 MarinOpenAI
1169±31316
21–29 KatieCartesia
1169±34128
31–34 DanielleAmazon Polly
1163±26826
41–38 CelesteRime
1143±36116
51–34 JonElevenLabs
1130±28926
61–36 AdrianFish Audio
1127±31725
71–34 KoreGoogle Gemini TTS
1112±26232
81–35 AshleyInworld
1109±27525
91–34 JoleneCartesia
1103±28126
101–37 HenrySpeechify
1092±30525
111–38 Spuds OxleyElevenLabs
1092±25228
122–34 OrusGoogle Cloud TTS
1081±20936
132–35 GeorgeSpeechify
1072±22528
143–37 KelseySmallest.ai
1067±20634
151–38 CharonGoogle Cloud TTS
1061±28421
161–37 Expressive narratorMiniMax
1060±30925
172–37 ClementineRime
1041±22123
182–39 CoralOpenAI
1034±25627
192–40 AlgenibGoogle Gemini TTS
1023±30321
204–40 EldrinInworld
1011±24422
215–38 Daniel (M)Murf
1005±23228
224–39 LeoSmallest.ai
1003±21528
232–42 LaurenCartesia
1003±34326
243–40 CedarOpenAI
1001±25723
254–42 AchernarGoogle Gemini TTS
980±29725
266–42 Wayne (M)Murf
971±29624
2710–40 ThaliaDeepgram
965±19531
289–42 AthenaDeepgram
951±29529
295–42 PatrickAmazon Polly
948±26224
307–41 HannahFish Audio
945±23323
3110–41 ChelseaSmallest.ai
935±24528
328–42 SarahFish Audio
934±23727
339–41 AoedeGoogle Cloud TTS
931±23825
347–42 PeterElevenLabs
929±25924
356–42 AlaricInworld
929±25823
364–42 RuthAmazon Polly
920±36918
377–42 CaptivatingStorytellerMiniMax
918±31126
388–42 LyraRime
911±27523
3914–42 GeffenSpeechify
889±24827
4011–42 PatientManMiniMax
887±29323
4116–42 NatalieMurf
856±22123
4210–42 CoraDeepgram
856±38121
—— TaraOrpheuscollecting (8 so far)8
—— SavannahZonos3-Minicollecting (7 so far)7
—— LeoOrpheuscollecting (5 so far)5
—— AidenQwen3-TTScollecting (5 so far)5
—— af_bellaKokorocollecting (3 so far)3
—— ZoeOrpheuscollecting (3 so far)3
—— RyanQwen3-TTScollecting (3 so far)3
—— DefaultChatterboxcollecting (2 so far)2
—— am_michaelKokorocollecting (2 so far)2
—— WyattZonos3-Minicollecting (2 so far)2
—— af_heartKokorocollecting (1 so far)1

Listener question: “Which narration would you choose to continue listening to for this nonfiction book?” · 550 judgments from 42 listeners so far · 42 voices ranked; the rest are still collecting. Range = ranks consistent with the data at 95%. Methodology

Independent blind listening evaluation. Paid, vetted US listeners.
Methodology · Evaluated audio · Judgment data · Contact