sulabhkatiyar/trn-indcnfr-hi-pilot
trn-indcnfr-hi-pilot — Teacher-Output Cache for Ternary-ASR Distillation Append-only teacher-output cache used to distill a ternary Hindi ASR student. It stores, per audio clip, ONLY: row_id (a deterministic source-shard/position pointer), top-k (k=64) CTC logits (vocab id + log-prob per kept entry, blank always kept), selected encoder hidden states (last-3 blocks: layers 14 / 15 / 16), fp16. No audio and no ground-truth transcripts are stored or redistributed.… See the full description on the dataset page: https://huggingface.co/datasets/sulabhkatiyar/trn-indcnfr-hi-pilot.
trn-indcnfr-hi-pilot — Teacher-Output Cache for Ternary-ASR Distillation
Append-only teacher-output cache used to distill a ternary Hindi ASR student. It stores, per audio clip, ONLY:
row_id(a deterministic source-shard/position pointer),- top-k (k=64) CTC logits (vocab id + log-prob per kept entry, blank always kept),
- selected encoder hidden states (last-3 blocks: layers 14 / 15 / 16), fp16.
No audio and no ground-truth transcripts are stored or redistributed.
Provenance
- Teacher: `ai4bharat/indicconformer_stt_hi_hybrid_ctc_rnnt_large` — license MIT.
- Source corpora (audio & transcripts NOT redistributed here; used only to compute teacher outputs):
- `ai4bharat/Kathbath` (hindi) — CC-BY-4.0.
- `ai4bharat/Shrutilipi` (hindi) — CC-BY-4.0.
- Attribution: AI4Bharat.
License
CC-BY-4.0. Redistributing the derived teacher outputs (logits + hidden states + ids) is not redistribution of the source works; attribution to AI4Bharat is required.