← Speech-to-Text 
Modulate
Velma 2
Measured 2026-08-08
Run Velma 2 on Speko →One API, every model on this board.
WER · batch
4.4%
95% CI 2.8–6.2 · n=50
WER · streaming
Read speech
5.35%
Conversation
14.41%
Details
Read speech
n=1541 · FLEURS test+dev+train · vendor-direct
Conversation
12.0–17.3% · EdAcc n=283 · 60 speakers, 2 passes, 2026-10-01
2 of 566 readings came back empty, scored as all-deletion.
Finalize
1.21s
p50–p90 · US East · n=200 · vendor-direct · ±4%
Time to first token
1.48s
p50–p90 · US East · n=200 · vendor-direct · ±0%
Cost / min
$0.00083
Published streaming rate for English Fast, $0.05/hr. Batch is $0.025/hr. The $0.06/hr this row used to carry is the MULTILINGUAL row, a different product from the one this id names. Not a measurement.
Share this result
Embed the live badge
Markdown
[](https://benchmarks.speko.ai/stt/modulate-velma-2)
HTML
<a href="https://benchmarks.speko.ai/stt/modulate-velma-2"><img src="https://benchmarks.speko.ai/badge/stt/modulate-velma-2.svg" alt="Speko stt rank"></a>
URL
https://benchmarks.speko.ai/stt/modulate-velma-2