kazalbrur/Bangla-asr-fastconformer-116m-dialects

Update README.md

Update README.md

Update README.md

Scope the internal-suite comparison; cross-link BanglaBench-v1

Deprecate: point to kazalbrur/bangla-stt-fastconformer-116m-dialects

Show the leaderboard figure in the benchmark section

Add benchmark figures

Add BanglaBench-v1 independent benchmark section

Rename weights to match repo id (116m)

Update README.md

Update README.md

Align parameter counts to exact measured values (121.5M / 115.6M)

Map IndicVoices to canonical Hub dataset id in eval matrix

Update README.md

Clarify streaming limitation (offline-trained, not cache-aware)

Update README.md

Expand model card (rich format): details, training, usage, examples, citation

Remove base_model tag from frontmatter (Model tree); keep written attribution

Release ⚡ Srotoswini-Fast: Bangla regional dialects ASR (120M)

initial commit