whatmodel.sh / stt / Saarika v2.5
Saarika v2.5
↗ share
+ contribute
Saarika v2.5 is a closed-weights speech-to-text model from Sarvam AI, released May 2025.
Catalogue identity notes
Exact Sarvam batch model 'saarika:v2.5' (language_code bn-IN) per benchmark config. Distinct from streaming saaras:v3-realtime.
Model information sources
Score within the Open ASR English short-form WER (March 2026) comparison set only.
$ how it compares
The speech-to-text models ranked either side of Saarika v2.5, out of 176 with a score.
| # | model | score | where |
|---|---|---|---|
| 1 |  Muse Voice Transcribe 1.0 | 71±3 | api $0.0030/audio min |
| 2 |  Grok Voice Transcribe 2.0 | 70±3 | api $0.0017/audio min |
| 3 | Saarika v2.5 | 70±5 | comparison only |
| 4 |  Universal-3.6 Pro Realtime | 70±3 | api $0.0075/audio min |
| 5 |  Linden 1 (Agent STT) | 69±3 | api $0.0050/audio min |
see all speech-to-text models · find the ones that fit your machine
$ what it's good at
other tests1 not in a task · 92% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Bangla STT benchmark · BanSpeech 1,001-clip batch CER | 6.9% ✎ | █·▄··▄··▄·▄▄ beats 6/6 | 92% |
$ where to run it
No verified API or download recorded yet · add one
$ how the score was worked out · Bangla Speech-to-Text Benchma… 92%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 Bangla Speech-to-Text Benchmark v1.0 (Riverborn) | 1 | 92% |
| prior · pull toward the middle | 8% | |
| weighted average | = 70.3 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json