Skip to content

Pull requests: NVIDIA-NeMo/Speech

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

[TTS] Add audio masking to EasyMagpie TTS
#16176 opened Aug 31, 2026 by rlangman Collaborator Loading…
1 of 3 tasks
fix(asr): fix beam score normalization ASR
#16175 opened Aug 31, 2026 by naymaraq Collaborator Draft
8 tasks
fix(tts): apply bin_loss_scale warmup ramp in AlignerModel community-request TTS waiting-on-maintainers Waiting on maintainers to respond
#16171 opened Aug 30, 2026 by udsy19 Contributor Loading…
3 of 7 tasks
fix(audio): correct estimate/target routing into CombinedLoss's ASR loss audio community-request waiting-on-maintainers Waiting on maintainers to respond
#16169 opened Aug 29, 2026 by udsy19 Contributor Loading…
3 of 8 tasks
fix(asr): keep zero-length utterances inactive in batched MALSD beam search ASR community-request waiting-on-maintainers Waiting on maintainers to respond
#16162 opened Aug 28, 2026 by udsy19 Contributor Loading…
3 of 7 tasks
fix(speechlm2): port SALM MTP vLLM review hardening ASR CI common TTS
#16158 opened Aug 28, 2026 by pzelasko Collaborator 6/6 Loading…
fix(asr): guard val_text_pred in the hybrid EOU validation pass ASR community-request waiting-on-maintainers Waiting on maintainers to respond
#16151 opened Aug 27, 2026 by udsy19 Contributor Loading…
3 of 7 tasks
Add EasyMagpie request-time audio conditioning TTS
#16149 opened Aug 27, 2026 by vklimkov-nvidia Member Loading…
8 tasks done
feat(asr): enable mixed precision cache-aware streaming ASR community-request waiting-on-customer Waiting on the original author to respond
#16147 opened Aug 26, 2026 by AdityaBiranje Loading…
[TTS] allow null context similarity in cuts common
#16145 opened Aug 26, 2026 by paarthneekhara Collaborator Loading…
[ASR] add cache-aware streaming transformer ASR
#16143 opened Aug 26, 2026 by stevehuang52 Collaborator Loading…
feat(speechlm2): distill MTP heads with LK loss
#16140 opened Aug 25, 2026 by Slyne Contributor Draft
Qwen3 5 support ASR common
#16135 opened Aug 25, 2026 by nune-tadevosyan Collaborator Loading…
1 of 8 tasks
Stop ConcatDataset from re-sharding itself on every epoch common community-request waiting-on-maintainers Waiting on maintainers to respond
#16133 opened Aug 25, 2026 by VaggelisGian Loading…
3 of 8 tasks
Honor minimum numba version in support checks community-request core Changes to NeMo Core waiting-on-maintainers Waiting on maintainers to respond
#16132 opened Aug 25, 2026 by VaggelisGian Loading…
5 of 8 tasks
Fix crash with Lhotse train data and float limit_train_batches ASR community-request waiting-on-maintainers Waiting on maintainers to respond
#16131 opened Aug 25, 2026 by VaggelisGian Loading…
3 of 8 tasks
Add DFlash speculative decoding support to NeMo SpeechLM vLLM plugin ASR
#16129 opened Aug 25, 2026 by Slyne Contributor Draft
5 of 6 tasks
ProTip! Add no:assignee to see everything that’s not assigned.