Outdated FLEURS audio dataset link in MMS ASR tutorial notebook - Githubissues

facebookresearch / fairseq

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

MIT License

30.23k stars 6.38k forks source link

Outdated FLEURS audio dataset link in MMS ASR tutorial notebook #5514

Open yongzx opened 3 months ago

yongzx commented 3 months ago

🐛 Bug

To Reproduce

Steps to reproduce the behavior (always include the command you ran):

Error in section 3: "Prepare Audio File" of MMS ASR tutorial notebook examples/mms/asr/tutorial/MMS_ASR_Inference_Colab.ipynb

Run command wget 'https://datasets-server.huggingface.co/assets/google/fleurs/--/en_us/train/0/audio/audio.mp3'
Error message of 403 Forbidden

Code sample

wget 'https://datasets-server.huggingface.co/assets/google/fleurs/--/en_us/train/0/audio/audio.mp3'
# HTTP request sent, awaiting response... 403 Forbidden

Expected behavior

It should download the FLEURS dataset. The bug is due to the outdated dataset link on HuggingFace.

Environment

fairseq Version (e.g., 1.0 or main): main
PyTorch Version (e.g., 1.0): 2.3.1
OS (e.g., Linux): Linux
How you installed fairseq (pip, source): source
Build command you used (if compiling from source): pip3 install --editable ./
Python version: 3.10.12
CUDA/cuDNN version: 12.1
GPU models and configuration:
Any other relevant information:

Additional context