Datasets for Urdu about xlsr-wav2vec2-speech-recognition-urdu HOT 5 CLOSED

@abdulwadood10 I have added the transcripted file.

from xlsr-wav2vec2-speech-recognition-urdu.

Thanks for that. Although I ran through your code with different csv file and there is one more issue..

model = Wav2Vec2ForCTC.from_pretrained("/content/gdrive/MyDrive/wav2vec2-urdu/checkpoint-800").to("cuda") processor = Wav2Vec2Processor.from_pretrained("/content/gdrive/MyDrive/wav2vec2-urdu/checkpoint-800")

After running through this code I face this error

can you point me where this "/content/gdrive/MyDrive/wav2vec2-urdu/checkpoint-800" file or link coming from? I am almost finished running with your code.

from xlsr-wav2vec2-speech-recognition-urdu.

This path is for model saving in the drive after certain checkpoints. You can add your own drive path or you can remove it completely if you do not want to save your model after a certain period during training.

from xlsr-wav2vec2-speech-recognition-urdu.

Alright, so this is saving the model on your own huggingface drive right?

What if I want to save it on my own local machine so I can run it offline.

I will be giving my folder path to the to 'pre_trained' parameter?

from xlsr-wav2vec2-speech-recognition-urdu.

I believe that you can not save it on your local machine. what you can do is, save it in the drive and download it later to use it on the local machine.

from xlsr-wav2vec2-speech-recognition-urdu.