fairseq

fairseq copied to clipboard

About extracting the feature of the 12th layer for finetuned hubert model

4

#### What is your question? In examples/hubert/simple_kmeans, there are methods of extracting the feature of _pretrained_ model. Now, for comparing the similarity of different models and layers, I am working...

LYPinASR

A question for dict.ltr.txt and dict.km.txt in Hubert

1

## ❓ Questions and Help ### Before asking: 1. search the issues. 2. search the docs. #### What is your question? I am running the code of Hubert on Librispeech...

Ther-nullptr

## ❓ Questions and Help ### Before asking: 1. search the issues. 2. search the docs. #### What is your question? #### Code #### What have you tried? #### What's...

lucky-asia

Hello, First of all, thanks for the great work on this project! I am looking to finetune the Wav2Vec model using fairseq-hydra-train on Intel accelerators. Could you provide some guidance...

vijethk-intel

## ❓ Questions and Help #### I followed the instructions from DrNMT to train a base MT model(fconv) first, and then hoped to use the MT model to prepare three...

123THU

question about mms tts text input

1

I am using mms-tts for speech synthesis in Hakka or Min Nan languages. I tried it out on the demo page at https://huggingface.co/spaces/mms-meta/MMS. However, when I input text in Chinese...

zhoujxwilliam

There seems to be an issue with the Adafactor optimizer found here, if beta1 is > 0: https://github.com/facebookresearch/fairseq/blob/ecbf110e1eb43861214b05fa001eff584954f65a/fairseq/optim/adafactor.py#L66 Please find a detailed description here: https://github.com/huggingface/transformers/issues/34506

dxqbYD

Do I need to crop long audio for inference based on pretrained models？

1

I have a 5-minute audio file, and the wav2vec features obtained by direct inference and the wav2vec features obtained by cropping into a 10s segment are inconsistent. Is it possible...

xuduo18311199384

Include Kirundi (Rundi - run_Latn) language in the No Language Left Behind (NLLB) project

1

## 🚀 Feature Request Include Kirundi (Rundi) language in the No Language Left Behind (NLLB) project ### Motivation > As researchers and developers from Burundi, we are deeply concerned about...

labKnowledge