AI model rankings

Best AI voice models

Text-to-speech models ranked by blind votes on which voice sounds more natural.

Updated Oct 30, 2026, 01:41 PM UTC Data from Artificial Analysis

#1 right now

Eleven v4 Turbo

ElevenLabs

1,336Elo

Runner-up

Eleven v4

ElevenLabs

1,334Elo · 2 behind

Ranked

+38

100models from 43 companies

Showing 100 of 100 models

Top 10 for text to speech

Elo rating from blind head-to-head votes. Higher is better. Click a bar to see the model.

Compare all text to speech models

100 models, ranked by Elo · click a model to see its page

Rank	Model	Elo	± range	Votes	Released
🥇	Eleven v4 Turbo ElevenLabs	1,336	-16/16	–	–
🥈	Eleven v4 ElevenLabs	1,334	-15/15	–	–
🥉	Qwen-Audio-3.1-TTS-Plus Alibaba	1,305	-15/15	–	–
4	Sonic 3.6 Cartesia	1,296	-13/13	–	–
5	Gemini 3.8 Flash TTS Google	1,283	-13/13	–	–
6	Fish Audio Drama 3 Fish Audio	1,278	-18/18	–	–
7	Qwen-Audio-3.0-TTS-Plus Alibaba	1,269	-13/13	–	–
8	Realtime TTS-2 Inworld	1,267	-14/14	–	–
9	Lightning V3.1 Pro (Oct 2026) Smallest.ai	1,266	-16/16	–	–
10	Gemini 3.8 Flash-Lite TTS Google	1,259	-13/13	–	–
11	Simba 3.2 SpeechifyAI	1,249	-12/12	–	–
12	Fish Audio Drama 3 Preview Fish Audio	1,240	-14/14	–	–
13	Breeze TTS 2 BreezeBlue	1,228	-13/13	–	–
14	AssemblyAI-TTS-1 AssemblyAI	1,227	-16/16	–	–
15	Voxtral 3 TTS Mistral	1,223	-14/14	–	–
16	Realtime TTS-2 Flash Inworld	1,217	-12/12	–	–
17	v3 Conversational ElevenLabs	1,215	-12/12	–	–
18	StepAudio 2.5 TTS (Aug 2026) StepFun	1,214	-13/13	–	–
19	Gemini 3.1 Flash TTS Google	1,212	-11/11	–	–
20	Luna TTS VUI Labs	1,205	-11/11	–	–
21	Sonic 3.5 Cartesia	1,194	-11/11	–	–
22	Fish Audio S2.1 Pro (Oct 2026) Fish Audio	1,188	-13/13	–	–
23	Eleven v3 ElevenLabs	1,188	-10/10	–	–
24	Ming-3.0-tts Ant Group	1,187	-13/13	–	–
25	Soniox TTS Real-Time v2 Soniox	1,182	-10/10	–	–
26	Gradium TTS (Sep 2026) Gradium	1,175	-13/13	–	–
27	Speech 2.8 HD MiniMax	1,174	-10/10	–	–
28	SpaceXAI TTS SpaceXAI	1,157	-13/13	–	–
29	Falcon 2 Murf AI	1,156	-12/12	–	–
30	Gradium TTS (Aug 2026) Gradium	1,154	-12/12	–	–
31	Speech 2.8 Turbo MiniMax	1,153	-10/10	–	–
32	Azure HD 2.5 Microsoft	1,145	-11/11	–	–
33	Async Flash v1.5 async	1,141	-11/11	–	–
34	Step TTS 2 (Mar 2026) StepFun	1,137	-11/11	–	–
35	Async Pro v1.0 async	1,129	-12/12	–	–
36	Speech 2.6 HD MiniMax	1,127	-11/11	–	–
37	Speech 2.6 Turbo MiniMax	1,122	-10/10	–	–
38	Simba 3.0 SpeechifyAI	1,117	-11/11	–	–
39	Chatterbox HD Resemble AI	1,110	-12/12	–	–
40	Fish Audio S2 Pro Fish Audio	1,109	-11/11	–	–
41	Speech-02-HD MiniMax	1,108	-11/11	–	–
42	TTS-1 HD OpenAI	1,106	-11/11	–	–
43	Step Audio EditX (Mar 2026) StepFun	1,099	-12/12	–	–
44	Speech 2.6 Hithink	1,096	-12/12	–	–
45	Turbo v2.5 ElevenLabs	1,095	-10/10	–	–
46	Multilingual v2 ElevenLabs	1,093	-10/10	–	–
47	Speech-02-Turbo MiniMax	1,091	-11/11	–	–
48	Voxtral TTS Mistral	1,090	-12/12	–	–
49	TTS-1 OpenAI	1,089	-10/10	–	–
50	OpenAudio S1 Fish Audio	1,087	-11/11	–	–
51	Gemini 2.5 Flash Lite TTS Google	1,085	-11/11	–	–
52	Flash v2.5 ElevenLabs	1,077	-10/10	–	–
53	Journey Google	1,074	-12/12	–	–
54	GPT-Realtime-2 OpenAI	1,073	-12/12	–	–
55	Studio Google	1,073	-11/11	–	–
56	T2A-01-HD MiniMax	1,072	-11/11	–	–
57	Gemini 2.5 Flash TTS (Dec 2025) Google	1,068	-12/12	–	–
58	Magpie-Multilingual 357M (Feb 2026) NVIDIA	1,067	-12/12	–	–
59	Polly Generative Amazon	1,063	-11/11	–	–
60	Maya 2 Flash Maya Research	1,062	-13/13	–	–
61	Chirp 3: HD Google	1,062	-11/11	–	–
62	Kokoro 82M v1.0 Kokoro	1,062	-11/11	–	–
63	Octave 2 Hume AI	1,060	-12/12	–	–
64	Sonic 3 Cartesia	1,056	-11/11	–	–
65	MiMo-V2.5-TTS Xiaomi	1,055	-12/12	–	–
66	Coda Rime	1,054	-12/12	–	–
67	Flux TTS Deepgram	1,052	-13/13	–	–
68	Async Flash v1.0 async	1,048	-11/11	–	–
69	Maya 2 Global Maya Research	1,048	-13/13	–	–
70	Maya1 Maya Research	1,046	-12/12	–	–
71	OpenAudio S1 Mini Fish Audio	1,044	-20/20	–	–
72	Gemini 2.5 Pro (Dec 2025) Google	1,042	-12/12	–	–
73	Octave TTS Hume AI	1,040	-11/11	–	–
74	Higgs Audio V3 TTS Boson AI	1,039	-13/13	–	–
75	Bland Speech v3 Bland AI	1,035	-12/12	–	–
76	Azure Neural Microsoft	1,033	-20/20	–	–
77	MAI-Voice-1 Microsoft	1,030	-12/12	–	–
78	T2A-01-Turbo MiniMax	1,028	-11/11	–	–
79	Polly Long-Form Amazon	1,027	-13/13	–	–
80	Chatterbox Resemble AI	1,024	-11/11	–	–
81	Lightning v3.1 Smallest.ai	1,022	-12/12	–	–
82	Raon SpeechLM Krafton	1,007	-13/13	–	–
83	Zonos-v0.1 Zyphra	1,000	0/0	–	–
84	Murf Speech Gen 2 Murf AI	986	-12/12	–	–
85	OpenVoice v2 OpenVoice	956	-13/13	–	–
86	VibeVoice 1.5B Microsoft	955	-14/14	–	–
87	Neuphonic TTS Neuphonic	941	-14/14	–	–
88	Qwen3 TTS Flash Alibaba	941	-14/14	–	–
89	Qwen3 TTS Alibaba	929	-13/13	–	–
90	Kugel 3 KugelAudio	925	-13/13	–	–
91	WaveNet Google	919	-12/12	–	–
92	XTTS v2 Coqui	915	-15/15	–	–
93	Mist V2 Rime	903	-15/15	–	–
94	Polly Neural Amazon	895	-14/14	–	–
95	Neural2 Google	894	-12/12	–	–
96	StyleTTS 2 StyleTTS	891	-15/15	–	–
97	Standard Google	888	-12/12	–	–
98	Noiz TTS Noiz	869	-15/15	–	–
99	MetaVoice v1 MetaVoice	847	-18/18	–	–
100	Polly Standard Amazon	832	-15/15	–	–

What do these numbers mean?

Elo rating: The same system used to rank chess players. People see two results for the same prompt without knowing which model made them, and vote for the better one. Winning against strong models raises the score more.
± range: How certain the score is. A range of −8/+8 means the true score is very likely within 8 points either way. Models whose ranges overlap are roughly tied.
Votes: How many head-to-head comparisons the model has been in. More votes means a more reliable rating.
Is #1 always best for me?: Not always. Small Elo gaps are hard to notice. Also check price, speed and whether the model is available in the app you use.

Benchmark data from Artificial Analysis, an independent benchmarking company. Rankings refresh every few hours.