Search results

TDT Head Stagnation in parakeet-tdt_ctc-110m Fine-tuning on Persian Data

Status: Open.

#14140 In NVIDIA/NeMo;

· saeedzou opened on Jul 6, 2025

The 2.4.0.rc1 image

Status: Open.

#14138 In NVIDIA/NeMo;

· XChen-Zero opened on Jul 6, 2025

When running diffusion, how should we handle cases where the sequence length cannot be evenly divided by the context parallelism (CP) size?

Status: Open.

#14137 In NVIDIA/NeMo;

· njw1123 opened on Jul 6, 2025

404/broken link http://www.speech.cs.cmu.edu/databases/an4/

Status: Open.

#14136 In NVIDIA/NeMo;

· FireballDWF opened on Jul 6, 2025

Information on finetuning the parakeet-tdt-0.6b-v2model for new languages!!

Status: Open.

#14132 In NVIDIA/NeMo;

· gkiri opened on Jul 5, 2025

speech_to_text_buffered_infer_rnnt.py why not support tiemstamps？

Status: Open.

#14119 In NVIDIA/NeMo;

· 1091492188 opened on Jul 3, 2025

I found that 11 minutes of audio occupies 20G video memory, how to solve it, using parakeet-tdt-0.6b-v2

bugSomething isn't workingSomething isn't working

Status: Open.

#14111 In NVIDIA/NeMo;

· 1091492188 opened on Jul 2, 2025

transcribe failed

bugSomething isn't workingSomething isn't working

Status: Open.

#14107 In NVIDIA/NeMo;

· wen1q84 opened on Jul 2, 2025

[Question] Is nemo support Lora moe part with ETP!=TP?

Status: Open.

#14085 In NVIDIA/NeMo;

· ShareLer opened on Jul 1, 2025

Help! Couldn't find Qwen3 pretraining recipes in the docker image!

Status: Open.

#14073 In NVIDIA/NeMo;

· tjoymeed opened on Jun 30, 2025

When pretraining using Nemo-Megatron, are the input data truncated and discarded?

Status: Open.

#14057 In NVIDIA/NeMo;

· tjoymeed opened on Jun 28, 2025

Bug: pad_to_seq_length uses incorrect condition when padding loss_mask

bugSomething isn't workingSomething isn't working

Status: Open.

#14041 In NVIDIA/NeMo;

· JackJessada opened on Jun 27, 2025