Search results

Question about the prompt format in nemotron-3.5-asr-streaming-0.6b

community-request

Status: Open.

#15784 In NVIDIA-NeMo/NeMo;

· MahmoudAshraf97 opened on Jun 10, 2026

nemotron-3.5-asr-streaming-0.6b.nemo: Each fine-tuning epoch uses only 1000 data

bugSomething isn't workingSomething isn't working

community-request

Status: Open.

#15782 In NVIDIA-NeMo/NeMo;

· ican24 opened on Jun 10, 2026

Embedding size mismatch when loading magpie_tts_multilingual_357m

bugSomething isn't workingSomething isn't working

community-request

Status: Open.

#15780 In NVIDIA-NeMo/NeMo;

· unilight opened on Jun 10, 2026

Fine-tuning FastConformer Arabic model with custom tokenizer results in WER=1.0 and no learning

bugSomething isn't workingSomething isn't working

community-request

waiting-on-customerWaiting on the original author to respondWaiting on the original author to respond

Status: Open.

#15778 In NVIDIA-NeMo/NeMo;

· Firaskraiem1 opened on Jun 9, 2026

model.transcribe() is not thread-safe: encoder.freeze()/unfreeze() race causes ValueError under concurrent inference

community-request

Status: Open.

#15771 In NVIDIA-NeMo/NeMo;

· beastoin opened on Jun 9, 2026

Parakeet TDT decodes empty when short utterance has trailing silence treated as valid audio

community-request

waiting-on-maintainersWaiting on maintainers to respondWaiting on maintainers to respond

Status: Open.

#15757 In NVIDIA-NeMo/NeMo;

· andimarafioti opened on Jun 5, 2026

decoding results aboult nvidia/parakeet-unified-en-0.6b

bugSomething isn't workingSomething isn't working

community-request

waiting-on-maintainersWaiting on maintainers to respondWaiting on maintainers to respond

Status: Open.

#15746 In NVIDIA-NeMo/NeMo;

· kobenaxie opened on Jun 3, 2026

[RFC] Unified vLLM-style Execution Chunking & State Caching for ASR Streaming Pipelines

community-request

waiting-on-customerWaiting on the original author to respondWaiting on the original author to respond

Status: Open.

#15740 In NVIDIA-NeMo/NeMo;

· shivansh023023 opened on Jun 1, 2026

[Bug] NGramGPULanguageModel.from_arpa() fails with AssertionError on valid 6-gram ARPA files

Status: Open.

#15715 In NVIDIA-NeMo/NeMo;

· amargolin78 opened on May 21, 2026

support max_speakers configuration in diar_streaming_sortformer_4spk-v2.1

community-request

waiting-on-maintainersWaiting on maintainers to respondWaiting on maintainers to respond

Status: Open.

#15711 In NVIDIA-NeMo/NeMo;

· livefantasia opened on May 19, 2026

Release >=v2.6.2 break pytorch lightning checkpoint callback when save_best_model=True

bugSomething isn't workingSomething isn't working

community-request

waiting-on-maintainersWaiting on maintainers to respondWaiting on maintainers to respond

Status: Open.

#15709 In NVIDIA-NeMo/NeMo;

· MahmoudAshraf97 opened on May 17, 2026

Relax lightning <= 2.4.0 cap to allow >= 2.6.1 (security fix for LightningModule.load_from_checkpoint deserialization)

bugSomething isn't workingSomething isn't working

community-request

waiting-on-maintainersWaiting on maintainers to respondWaiting on maintainers to respond

Status: Open.

#15706 In NVIDIA-NeMo/NeMo;

· GiliBekhor opened on May 17, 2026