Artificial Curiosity Labs

Speech models

Speech recognition, text-to-speech, and speech-to-speech quality.

Snapshot: August 4, 2026

Speech-to-text

Artificial Analysis WER Index · Lower is better

Eligible
0
Candidates
63
Coverage
0.0%

unavailable

The Free API rounds AA-WER too coarsely to produce a reliable ranking for this snapshot.

Text-to-speech

Text-to-speech Elo · Higher is better

Eligible
92
Candidates
92
Coverage
100.0%
Rank Model Creator Text-to-speech Elo Since prior snapshot
1 Simba 3.2 SpeechifyAI 1,229 Improved 1
2 Qwen-Audio-3.0-TTS-Plus Alibaba 1,227 Down 1
3 Gemini 3.1 Flash TTS Google 1,212 No change

Speech-to-speech BBA

BBA score · Higher is better

Eligible
33
Candidates
37
Coverage
89.2%
Rank Model Creator BBA score Since prior snapshot
1 tie Qwen3.5 Omni Plus Realtime Alibaba 0.99 No change
1 tie Qwen Audio 3.0 Realtime Plus Alibaba 0.99 No change
3 Step-Audio R1.1 (Realtime) StepFun 0.98 No change

Speech-to-speech FDB

FDB score · Higher is better

Eligible
25
Candidates
37
Coverage
67.6%
Rank Model Creator FDB score Since prior snapshot
1 Qwen Audio 3.0 Realtime Plus Alibaba 0.98 No change
2 Qwen Audio 3.0 Realtime Flash Alibaba 0.97 No change
3 tie GPT-Realtime-2.1 High OpenAI 0.96 No change
3 tie GPT-Realtime-2 Minimal OpenAI 0.96 No change
3 tie GPT-Realtime-1.5 OpenAI 0.96 No change
3 tie GPT Realtime Mini (Oct 2025) OpenAI 0.96 No change

Speech-to-speech Tau Voice

Tau Voice score · Higher is better

Eligible
20
Candidates
37
Coverage
54.1%
Rank Model Creator Tau Voice score Since prior snapshot
1 Grok Voice Think Fast 2.0 High SpaceXAI 0.56 Not comparable
2 Qwen Audio 3.0 Realtime Plus Alibaba 0.55 Down 1
3 Grok Voice Think Fast 1.0 SpaceXAI 0.52 Down 1

Data source: Artificial Analysis .