Views
No views yet
nvidia/nemotron-3.5-asr-streaming-0.6b
on a vast.ai H100. Repo of record (scripts, RECIPES.md, STATUS.md): git
nt-3.5-asr-streaming-ft. Repo was private, made public 2026-07-05
(private-storage quota).checkpoints/2026-07-05_03-18-46/nemotron35_he_mix_ct5--val_wer=0.2085-epoch=198.ckpt
(= nemo_models/ckpt_mix_final.nemo, weights-only 2.4G; the leg .nemo backup
in the same dir is identical). Mixed run: committees 0.5 / CT-v5 0.5 (Lhotse
input_cfg), 200k steps total, init from ckpt150k (committees-only run 2).
Final global-best val_wer 0.20851@199k (CT-v5 dev, canonical Hebrew).checkpoints/2026-07-05_03-18-46/...epoch=198-last.ckpt
(fp32 + Adam moments + trainer state, 7.7G) via
+exp_manager.resume_from_checkpoint=... — see RECIPES.md §4.
Training data: committees tars in HF dataset
benderrodriguez/knesset-committees-prepped; CT-v5 tars NOT archived
(gated source) — re-prep with scripts/prepare_ct5_train.py + scripts_box/prep_ct5_box.sh.| benchmark | base | com.@20k | @40k | @50k | @150k | mix@49k | @79k | @100k | mix final @199k |
|---|---|---|---|---|---|---|---|---|---|
| CT-v5 test (2000) | 0.3570 | 0.3058 | 0.2792 | 0.2671 | 0.2733 | — | 0.1947 | 0.1906 | 0.1806 |
| committees dev (946) | 0.4620 | 0.3391 | 0.3196 | 0.3135 | 0.2959 | — | 0.2871 | 0.2918 | 0.2826 |
| eval-d1 (17 long-form) | 0.1892 | 0.1911 | 0.1713 | 0.1700 | 0.1543 | 0.1394 | 0.1393 | 0.1367 | 0.1374 |
| eval-whatsapp (54) | 0.3273 | — | — | — | 0.2658 | 0.2062 | 0.2011 | 0.1994 | 0.1998 |
outputs/ with scripts/score_canonical.py (identical test subsets
throughout; earlier notes mislabeled ckpt50k results as "base").scripts/score_canonical.py; predictions in outputs/eval_*.checkpoints/<run|leg>/ — best + -last Lightning .ckpt per leg (7.7G each)nemo_models/ — deployable .nemo milestones (ckpt20k/40k/50k/150k,
ckpt_mix49k/79k/100k, ckpt_mix_final)nemo_experiments_meta/ — full exp tree w/o ckpt/nemo: TensorBoard events,
hparams.yaml, cmd-args for every leglogs/ — all console logs (train legs incl. crash-preserved, evals, uploads)outputs/ — eval prediction manifests for every eval runscripts_box/ — box-only scripts not in gitMANIFEST_all_ckpts_on_box.txt — ls -laR at first-archive time