Catalog · 40 models · 13 providers
Every speech-to-text model worth using, one rate card.
40
Total models
106
Languages
$0.004
Cheapest /min
—
Best WER
2 models
filtered from 40
ElevenLabs
State-of-the-art batch STT — 90+ languages, speaker diarization, audio tagging
Speed Factor
0.2x
Uptime
Billing
per 1s
Retention
76 languages
(view all)
ElevenLabs
Most accurate low-latency STT — <150ms, 90+ languages
TTFW (P50)
Speed Factor
0.0x
Billing
per 1s
Retention
76 languages
(view all)