fairseq
fairseq copied to clipboard
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
Hi. My goal is to finetune a large BERT-based MT model (e.g. `NLLB-200-1.3B`) on new words that are out of model's vocabulary. I managed to finetune it only from a...
As a ton of people don't know how to use NLLB project here is a short tutorial to get it running directly through fairseq! 1. Follow the installation instructions [here](https://github.com/facebookresearch/fairseq/blob/nllb/INSTALL.md)....
In the same line of thoughts as the MMS ASR finetuning, I've created a [repo](https://github.com/ylacombe/finetune-hf-vits) that allows MMS TTS finetuning, and updated the MMS README to reflect it! You can...
## ❓ Questions and Help ### Before asking: 1. search the issues. 3. search the docs. I searched the issues and the docs. It seems no one mentioned this question...
## 🐛 Bug wav2vec (the og one) from the [wav2vec readme]( https://github.com/facebookresearch/fairseq/blob/main/examples/wav2vec/README.md#wav2vec), when called, as I understand, should return the contrastive predictive logits from the paper that are used to...
## 🐛 Bug ### To Reproduce I followed the steps in the simul_mustc documentation step by step, but I found that the accuracy of training the ASR model was not...
# Before submitting - [ ] Was this discussed/approved via a Github issue? (no need for typos, doc improvements) - [x] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)? - [ ]...
## 🚀 Feature Request I hope there could be a code base for videoclip with newer version of dependencies (such as python, pytorch, transformers, diffusers, etc) ### Motivation Recently I...
## ❓ Questions and Help ### Before asking: This issue should be mentioned in data2vec v2 paper explicitly, instead of roughly explane in few phase. So, there have no sufficient...
[ASR] Is lexicon-free W2lKenLMDecoder beam search decoding possible with no external language model?
Hello, I am wondering whether it is possible to run lexicon-free beam search decoding using W2lKenLMDecoder **without** an external language model? I am trying to initialise my decoder with the...