Using Songs to Improve Kazakh Automatic Speech Recognition
research work·conference paper·active
LREC 2026 paper evaluating sung Kazakh data as a complementary source for low-resource ASR.
Recorded facts
| Official site | https://arxiv.org/abs/2603.00961 ↗ |
|---|---|
| Geography | Kazakhstan / Kazakh language |
| doi | 10.63317/5hqonmz5roum |
| arxiv | 2603.00961 |
| venue | LREC 2026 |
| author | Rustem Yeshpanov |
| dedupe risk | low |
| discovery route | multilingual primary-source discovery |
| research topics | sung speech; Kazakh ASR; domain adaptation |
| proposed subtype | conference_paper |
| evidence strength | strong |
| recommended prompt | 06-research-works |
| proposed entity type | ResearchWork |
| classification rationale | Material AI-music, audio-AI, MIR, dataset, or music-rights relevance is explicit in the cited source. |
Current
| introduces | Kazakh Songs ASR DatasetSource 1 ↗ VERIFIED high confidence |
|---|
Sources & changes
Checked 10d ago · highhow verification works
Field-level evidence
Public change history
- Status unknown → active
- Official URL unknown → arxiv.org/abs/2603.00961
- Record maintenance · 12 fields updated
Is this yours? Claim this record →·See something wrong? Report a correction →