How did you get the audio for "datasets/srcdata/msrvtt/audios"?

[wonzin]

The original msrvtt folder structure is the below.

msrvtt
├── annotation
│ ├── MSR_VTT.json
├── high-quality
│ ├── structured-symlinks
│ │ ├── jsfusion_val_caption_idx.pkl
│ │ ├── ... many other files....
├── structured-symlinks
│ ├── jsfusion_val_caption_idx.pkl
│ ├── ... many other files....
├── videos
│ ├── all
│ │ ├── video1.mp4
│ │ ├── ....
│ │ ├── video9999.mp4
│ ├── tmp
│ │ ├──MSRVTT.zip
│ ├── vids
│ │ ├──data
│ │ │ ├── MSRVTT.zip

However, there is no audios for msrvtt.

1. How did you get the audio?
   Is there specific way to extract the audio for example, bitrate, sample rate, audio channel, type of codec.
   Any kind of audio file is valid?

2. "datasets/src/data/msrvtt/videos" == "msrvtt/videos/all" ?

[wonzin]

ffmpeg video.mp4 -ac 1 -ar 16000 audio.wav
I use this options to convert into audios. but it still has other error.

04/15/2024 14:39:07 - INFO - main - data_cfg_msrvtt_cap_val_batch_size : 64
04/15/2024 14:39:07 - INFO - main - msrvtt_cap Using clip mean and std.
04/15/2024 14:39:07 - INFO - main - msrvtt_cap transforms crop_flip
04/15/2024 14:39:07 - INFO - main - Create Dataset msrvtt_cap Success
04/15/2024 14:39:07 - INFO - main - loader cap%tvas--msrvtt_cap , ratio 10170 , bs_pergpu 16, n_workers 8

not have audios video6446
not have audios video51
...

[DelusionalLogic]

I've used the script in https://github.com/TXH-mercury/VAST/blob/410ca47acf40d4ab098e345b76159df66bc42239/utils/offline_process_data.py to extract the audio.

As for still getting no audio errors. I think some of the videos don't contain any audio, and those errors are expected.

[wonzin]

Thank you you save my day :)