fairseq

fairseq copied to clipboard

I think there are errors in the following lines: https://github.com/facebookresearch/fairseq/blob/7409af7f9a7b6ddac4cbfe7cafccc715b3c1b21e/fairseq/optim/amp_optimizer.py#L23 I guess should be `amp_kwargs = {"init_scale": cfg.common.amp_init_scale}` https://github.com/facebookresearch/fairseq/blob/7409af7f9a7b6ddac4cbfe7cafccc715b3c1b21e/fairseq/optim/amp_optimizer.py#L24-L25 I guess should be `amp_kwargs["growth_interval"] = cfg.common.amp_scale_window`

slikos

Hypo.word file not found

1

## question What I plan to use is https://colab.research.google.com/github/facebookresearch/fairseq/blob/main/examples/mms/asr/tutorial/MMS_ASR_Inference_Colab.ipynb  FileNotFoundError: [Errno 2] No such file or directory: '/tmp/tmphred1hfk/hypo.word' ## environment - fairseq Version main - Torch...

ahtelekb

#### What is your question? When I resynthesize audio using HuBERT, TTS and waveglow checkpoint I am not getting correct results. it seems like my generated audio is truncated. ####...

KaushalNaresh

# Before submitting - [ ] Was this discussed/approved via a Github issue? (no need for typos, doc improvements) - [ ] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)? - [...

hygong-fb

High Memory consumption on ASR inferencing

2

## ❓ Questions and Help #### What is your question? High memory consumption on ASR inferencing on a 16GB RAM server. It used up the system RAM up to 15.67GB...

olawale1rty

Translation fails when from mbart50 fine-tuned preloaded model

1

## 🐛 Bug I have used mbart50 pre-trained model to fine-tune a translation task for Si-En Languages. I now need to preload the model for translations. However during translation I...

aloka2209

The NLLB team describes in the 'No Language Left Behind: Scaling Human-Centered Machine Translation' (2022) paper experiments it did on adding different types of backtranslated data. Does anyone know whether...

miwytt

I want to train FaisrSeq Speech_to_text of my indic_timit_dataset. How can that be done

1

From the original Indic timit dataset I have created my own train, test, and validation dataset. The issue is the generation manifest, config files, vocabulary files, and feature bank zip....

SonakshiChauhan

## 🐛 Bug **Description**: The latest versions of Librosa (tested with 0.10.1) have introduced a required "size" keyword for the second argument of the `util.pad_center` function. This change causes an...

maureendss

DEPRECATION: omegaconf 2.0.6 has a non-standard dependency specifier PyYAML>=5.1.*. pip 24.0 will enforce

6

## 🐛 Bug Installing fairseq pops a deprecation notice related to its requirements. DEPRECATION: omegaconf 2.0.6 has a non-standard dependency specifier PyYAML>=5.1.*. pip 24.0 will enforce this behaviour change. A...

xvdp