38 7 11

Stephen Fernandes

StephennFernandes

StephennFernandes

AI & ML interests

Natural Language Processing , Reinforcement Learning

Recent Activity

new activity 2 days ago

google/gemma-3n-E4B-it:how to let the model transcribe long audio files ?

liked a dataset 22 days ago

MegaScience/MegaScience

liked a dataset about 1 month ago

microsoft/Updesh_beta

View all activity

Organizations

New activity in google/gemma-3n-E4B-it 2 days ago

how to let the model transcribe long audio files ?

➕ 1

#37 opened 2 days ago by

StephennFernandes

New activity in neavo/modern_bert_multilingual about 1 month ago

h-params used to pretrain from scratch

#8 opened about 1 month ago by

StephennFernandes

New activity in glaiveai/reasoning-v1-20m about 1 month ago

any plans on releasing the v2 of this dataset anytime soon ?

#7 opened about 1 month ago by

StephennFernandes

New activity in junnei/gemma-3-4b-it-speech about 2 months ago

AttributeError: 'Gemma3TextModel' object has no attribute 'prepare_inputs_for_generation'

#5 opened about 2 months ago by

StephennFernandes

New activity in nvidia/parakeet-tdt-0.6b-v2 about 2 months ago

how does the model handle timestamp decoding ?

👍 1

#50 opened about 2 months ago by

StephennFernandes

New activity in espnet/owsm_v4_medium_1B about 2 months ago

where can i find the v4 paper

#1 opened 2 months ago by

StephennFernandes

New activity in Dmitriy/bengali_with_clast 2 months ago

can you tell me the source.

#2 opened 2 months ago by

StephennFernandes

New activity in junnei/gemma-3-4b-it-speech 3 months ago

speech?

#1 opened 5 months ago by

MikuXvDev

New activity in microsoft/Phi-4-multimodal-instruct 4 months ago

Experience with Phi-4-Multimodal vs. Whisper-1 for Speech-to-Text

#39 opened 5 months ago by

hdevio

New activity in google/gemma-3-4b-it 4 months ago

I'm releasing the speech version of Gemma-3!

❤️ 👍 6

#35 opened 5 months ago by

junnei

New activity in facebook/w2v-bert-2.0 about 1 year ago

ValueError: negative dimensions are not allowed

#26 opened about 1 year ago by

StephennFernandes

How to use an LM like n-gram LM with w2v-bert-2.0?

#22 opened over 1 year ago by

lukarape

How to use n-gram with this model?

#24 opened over 1 year ago by

zkarapet00

New activity in nvidia/Llama3-ChatQA-1.5-8B over 1 year ago

Megatron LM training (fine-tuning) code ?

#9 opened over 1 year ago by

StephennFernandes

New activity in ai4bharat/sangraha over 1 year ago

how to download sangraha synthetic dataset ?

#3 opened over 1 year ago by

StephennFernandes

New activity in ylacombe/w2v-bert-2.0 over 1 year ago

ValueError: negative dimensions are not allowed

#13 opened over 1 year ago by

StephennFernandes

New activity in ai4bharat/sangraha over 1 year ago

datasets.utils.info_utils.NonMatchingSplitsSizesError:

#4 opened over 1 year ago by

StephennFernandes

New activity in openai/whisper over 1 year ago

Fine-tuning Whisper in more than one language

👍 3

#90 opened about 2 years ago by

andrespm

New activity in aaditya/orca_dpo_pairs-Hinglish-Codemix over 1 year ago

transliteration and translation model used ?

#3 opened over 1 year ago by

StephennFernandes

New activity in openai/whisper-large-v3 over 1 year ago

How to adapt for low resource language?

#73 opened over 1 year ago by

Imran1

Stephen Fernandes

AI & ML interests

Recent Activity

Organizations

StephennFernandes's activity

how to let the model transcribe long audio files ?

h-params used to pretrain from scratch

any plans on releasing the v2 of this dataset anytime soon ?

AttributeError: 'Gemma3TextModel' object has no attribute 'prepare_inputs_for_generation'

how does the model handle timestamp decoding ?

where can i find the v4 paper

can you tell me the source.

speech?

Experience with Phi-4-Multimodal vs. Whisper-1 for Speech-to-Text

I'm releasing the speech version of Gemma-3!

ValueError: negative dimensions are not allowed

How to use an LM like n-gram LM with w2v-bert-2.0?

How to use n-gram with this model?

Megatron LM training (fine-tuning) code ?

how to download sangraha synthetic dataset ?

ValueError: negative dimensions are not allowed

datasets.utils.info_utils.NonMatchingSplitsSizesError:

Fine-tuning Whisper in more than one language

transliteration and translation model used ?

How to adapt for low resource language?