NVIDIA-NeMo/Speech · Discussions

NVIDIA-NeMo · GitHub

Pinned Discussions

Licence of pretrained models

Q&A

DanBmh

NeMo ASR Demo for Transcription

Show and tell

titu1994

Run Hugging Face Models Instantly with Day-0 Support from NVIDIA NeMo Framework!

Show and tell

snowmanwwg

Sort by:Latest activity

Filter by label

Filter Use alt + click/return to exclude labels.

Categories, most helpful, and community links

Categories

LoadingCommunity links

Community guidelines

docs.nvidia.com/nemo/speech/nightly/index.html

Discussions

You must be logged in to vote

Is GPU-PB phrase boosting usable on Apple Silicon (MPS / CPU-only), or is it effectively NVIDIA-only?

zook8

asked

Aug 29, 2026 in

Q&A

· Unanswered

You must be logged in to vote

Built double-tap-Ctrl dictation for Linux on nemotron-speech-streaming-en-0.6b

NelsonScott

started

Aug 20, 2026 in

Show and tell

You must be logged in to vote

Adapting a model to output filler words and stutters

LahiLuk

asked

Sep 13, 2023 in

Q&A

· Unanswered

You must be logged in to vote

Define output language for multilingual nvidia/parakeet-tdt-0.6b-v3 model

rhenanbartels

asked

Sep 2, 2025 in

Q&A

· Unanswered

You must be logged in to vote

Run Parakeet TDT on Mac & Windows with one click — WhisperScript

jonathgh

started

Mar 12, 2026 in

Show and tell

You must be logged in to vote

SOS 🚨: Training ASR on Arabic/English data predicts blank token after ~100 steps and model is not learning.

Houss3m

asked

Aug 30, 2025 in

Q&A

· Unanswered

You must be logged in to vote

How to do context-prompting for generating a column in synthetic dataset?

Achmad96

asked

Feb 28, 2026 in

Q&A

· Unanswered

You must be logged in to vote

include word boundaries in the biasing

afaiza

asked

Jan 13, 2026 in

Q&A

· Unanswered

You must be logged in to vote

Issue with transcribing audio as a list when using timestamps

Filip858299

asked

Mar 11, 2026 in

Q&A

· Unanswered

You must be logged in to vote

Nemotron 3.5 ASR Streaming 0.6B in native C++/GGML. RTF = 0.0066 on RTX 5090

0xShug0

started

Jul 9, 2026 in

General

You must be logged in to vote

What If NeMo Had Actual Physics?

umaywant2

started

May 26, 2026 in

Ideas

You must be logged in to vote

About the NeMo Framework restructuring

lucacorbucci

asked

Mar 18, 2026 in

Q&A

· Unanswered

You must be logged in to vote

Repeat Pad vs Zero Pad for training speaker label models

CrypticFate5

asked

Mar 18, 2026 in

Q&A

· Unanswered

You must be logged in to vote

Transformers 5+ compatibility

BBC-Esq

started

Mar 15, 2026 in

Show and tell

You must be logged in to vote

Mnemo Cortex Memory

GuyMannDude

started

Mar 11, 2026 in

Show and tell

You must be logged in to vote

How does Nemotron CC & Nemo Curator "sort" its data?

XevWright

asked

Feb 24, 2026 in

Q&A

· Unanswered

You must be logged in to vote

[SpeechLM2] Bucketing training data

AudranBert

asked

Feb 5, 2026 in

Q&A

· Answered

You must be logged in to vote

pip install "nemo_toolkit[asr]"

grom311

started

Feb 23, 2026 in

General

You must be logged in to vote

Parakeet v2, v3 max_duration difference

mingewang

asked

Dec 18, 2025 in

Q&A

· Unanswered

You must be logged in to vote

(Parakeet V3) I can't convert the RNNTDecodingConfig timestamps into actual seconds

GreatHalf

asked

Dec 16, 2025 in

Q&A

· Answered

You must be logged in to vote

For Canary Varients for a AST task, would it be sufficient to train AST alone, or would ASR need to be trained

athaze

asked

Jan 9, 2026 in

Q&A

· Answered

You must be logged in to vote

Fine-tuned Arabic Quran ASR model works in NeMo but produces empty or incorrect transcriptions in NVIDIA Riva (streaming & offline)

NovaConAI

asked

Dec 18, 2025 in

Q&A

· Unanswered

You must be logged in to vote

NeMo ASR training sanity check: streaming RNNT+CTC setup, custom training_step, and input signature mismatch

NovaConAI

asked

Dec 17, 2025 in

Q&A

· Unanswered

You must be logged in to vote

How to provide multiple datasets with different weightage for llava next pre-training script?

gagangayari

started

Dec 13, 2025 in

General

You must be logged in to vote

Recipe Optimization Request: GPT-OSS 20B DAPT on 8x B200

bolivasNV

asked

Jan 13, 2026 in

Q&A

· Unanswered

…