Discussions

• 6 You must be logged in to vote

  🤖

  Report: Good Japanese ASR results with Fast Conformer

  fujimotos started Feb 21, 2024 in Public Models

  

  

  

  

  14
• 1 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

  

  

  0
• 1 You must be logged in to vote

  🙏

  Pretrained model refuses to converge.

  MostafaAhmed98 asked Apr 27, 2024 in Q&A · Unanswered

  

  

  1
• 1 You must be logged in to vote

  💬

  Finetuning a causal model from a non-causal checkpoint

  pointersbad started Aug 28, 2025 in General

  

  

  0
• 1 You must be logged in to vote

  🙏

  How to import qwen model from hf?

  RunjiaChen asked Aug 14, 2025 in Q&A · Unanswered

  

  

  1
• 3 You must be logged in to vote

  💬

  Finetune GPT2 model with TP=2

  AnirudhVIyer started Mar 13, 2024 in General

  

  

  

  

  2
• 1 You must be logged in to vote

  🙏

  nemo llm import fails with gpt_oss_120b

  DevHector-Bit asked Aug 20, 2025 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  💬

  How to preserve preserve gradient flow through preprocessor (AudioPreprocessor)?

  belson17 started Aug 12, 2025 in General

  

  

  0
• 1 You must be logged in to vote

  🙏

  ERROR: Make sure your audio length has enough samples for a single feature in 2nd epochs

  mehadi92 asked May 25, 2023 in Q&A · Unanswered

  

  

  3
• 1 You must be logged in to vote

  🙏

  Fine-tuning "parakeet-tdt_ctc" model "val_wer_ctc" is decreasing but "val_wer" is not deceasing

  mehadi92 asked Jul 29, 2025 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  💬

  If I use latest versions of nemo, also want to use python latest version with it.

  genagurbanguliyev started Jul 28, 2025 in General

  

  

  0
• 1 You must be logged in to vote

  💬

  Docker image size nvcr.io/nvidia/nemo:25.04 is 29 GB, is there smaller image i can use.

  chetandhembre started Jul 9, 2025 in General · Closed

  

  

  

  

  4
• 1 You must be logged in to vote

  🙏

  Finetuning IndicConformer -Speech to Text on custom data

  ananyagiliyal asked Jul 18, 2025 in Q&A · Unanswered

  

  

  0
• 0 You must be logged in to vote

  🙏

  Bug or a feature in streaming_utils.py BatchedFrameASRTDT? Is it possible to run parakeet-tdt-0.6b-v2 in streaming mode?

  SatenHarutyunyan asked May 29, 2025 in Q&A · Unanswered

  

  

  

  

  3
• 1 You must be logged in to vote

  🙏

  How to convert a large model(e.g., with hundreds of billions of parameters) from Hugging Face to NeMo format ？

  gaojingwei asked Jul 2, 2025 in Q&A · Unanswered

  

  

  2
• 1 You must be logged in to vote

  💬

  Training data pipeline for global brands (canary-1b)

  PardisTaghavi started Jun 27, 2025 in General

  

  

  

  

  2
• 3 You must be logged in to vote

  💬

  Why is checkpoint averaging being deprecated?

  riqiang-dp started Apr 8, 2025 in General

  

  

  2
• 1 You must be logged in to vote

  🙏

  Handling large amount of data where all of it is not present at any moment

  paniabhisek asked Jun 3, 2025 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙌

  Finetune the parakeet v2 tdt 0.6 b model for other language than English let's say Hindi

  deepanshu-yadav started Jun 2, 2025 in Show and tell

  

  

  0
• 1 You must be logged in to vote

  💬

  dataset features (merging, loading, streaming from remote)

  hawkoli1987 started May 30, 2025 in General

  

  

  0
• 1 You must be logged in to vote

  💬

  How to pretrain Qwen 2.5 7B with 128K context length?

  tjoymeed started May 26, 2025 in General

  

  

  1
• 1 You must be logged in to vote

  💬

  Does NeMo support dynamic YaRN?

  tjoymeed started May 26, 2025 in General

  

  

  0
• 1 You must be logged in to vote

  🙏

  [ASR] Begining of finetuning with lhotse worsen the base model

  AudranBert asked Feb 21, 2025 in Q&A · Unanswered

  

  

  4
• 1 You must be logged in to vote

  🙏

  Fine Tuning Cache Aware Streaming Model

  rohan25k asked Apr 29, 2024 in Q&A · Unanswered

  

  

  

  

  2
• 2 You must be logged in to vote

  🙌

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

  snowmanwwg started May 21, 2025 in Show and tell

  

  

  0