-
Notifications
You must be signed in to change notification settings - Fork 3.6k
Pull requests: NVIDIA-NeMo/Speech
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix(asr): apply the eou_pred filter instead of testing dict keys with hasattr
ASR
community-request
#16086
opened Aug 18, 2026 by
udsy19
Loading…
4 of 8 tasks
fix(tts): report phoneme dictionary entries dropped by the alphabet filter
community-request
TTS
#16084
opened Aug 18, 2026 by
udsy19
Loading…
3 of 7 tasks
fix(common): reach the max_open_streams guard instead of crashing on to_shards
common
community-request
#16082
opened Aug 18, 2026 by
udsy19
Loading…
3 of 7 tasks
fix(tts): guard trailing-space strip in German and Italian phoneme tokenizers
common
community-request
TTS
#16080
opened Aug 18, 2026 by
udsy19
Loading…
5 of 7 tasks
fix(asr): stop EOU segment end_time from drifting across utterances
ASR
community-request
#16078
opened Aug 18, 2026 by
udsy19
Loading…
5 of 8 tasks
fix: relax lightning cap to <2.7.0 to allow the load_from_checkpoint security fix
community-request
core
Changes to NeMo Core
#16076
opened Aug 18, 2026 by
udsy19
Loading…
5 of 8 tasks
fix(asr): make DALI window name validation reachable
ASR
community-request
#16075
opened Aug 18, 2026 by
udsy19
Loading…
4 of 8 tasks
fix(asr): accumulate chunk offsets in raw frames to stop timestamp drift
ASR
community-request
#16074
opened Aug 18, 2026 by
udsy19
Loading…
3 of 8 tasks
Improve MT-Parakeet streaming inference and timestamps
ASR
Speaker Tasks
#16070
opened Aug 17, 2026 by
ipmedenn
Collaborator
Loading…
6 of 8 tasks
Add Nemotron Transcribe target and MTP inference to SpeechLM vLLM
ASR
common
#16068
opened Aug 16, 2026 by
DongjiGao
Contributor
Loading…
fix(asr): preserve Unicode in transcription JSON
ASR
community-request
waiting-on-customer
Waiting on the original author to respond
#16066
opened Aug 15, 2026 by
sylvesterkaczmarek
Loading…
1 of 3 tasks
Derive word-confidence boundaries from the decoded text itself
ASR
community-request
waiting-on-maintainers
Waiting on maintainers to respond
#16064
opened Aug 13, 2026 by
Eoin-Houstoun
Loading…
Coalesce missing batch_size in ClusteringDiarizer dataloader configs
ASR
community-request
waiting-on-maintainers
Waiting on maintainers to respond
#16063
opened Aug 13, 2026 by
Eoin-Houstoun
Loading…
Remove PR babysitter automation, update commit/PR check instructions
CI
#16062
opened Aug 13, 2026 by
pzelasko
Collaborator
Loading…
Preserve valid codec subframe before audio EOS
TTS
#16059
opened Aug 13, 2026 by
vklimkov-nvidia
Member
Loading…
Fix variable name for audio file path in '_write_to_tar' function
community-request
waiting-on-maintainers
Waiting on maintainers to respond
#16058
opened Aug 13, 2026 by
advait-bm
Loading…
5 of 8 tasks
Add native THD packed ASR encoders
ASR
common
#16053
opened Aug 11, 2026 by
pzelasko
Collaborator
•
3/3
Loading…
Replace pointwise convolutions with matmul in ConformerEncoder: 10% E2E training speedup
ASR
community-request
waiting-on-maintainers
Waiting on maintainers to respond
#16052
opened Aug 11, 2026 by
MahmoudAshraf97
Contributor
Loading…
2 of 8 tasks
Support chunk microbatching and idxpack for multimodal SpeechLM data
common
#16048
opened Aug 10, 2026 by
KunalDhawan
Collaborator
•
2/3
Loading…
adding support for multilinugal cache aware model
ASR
#16047
opened Aug 10, 2026 by
naymaraq
Collaborator
Loading…
2 of 8 tasks
[Streaming SpeechLM] Add audomodel support
ASR
#16045
opened Aug 7, 2026 by
stevehuang52
Collaborator
•
Draft
add simulstream backend + qwen reasoning models for ST
ASR
#16043
opened Aug 7, 2026 by
lilithgrigoryan
Collaborator
Loading…
8 tasks
Previous Next
ProTip!
Updated in the last three days: updated:>2026-08-15.