# How to finetune whisper model

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-finetune-whisper-model%2F38986%3E%5C
**Category:** 🤗Transformers\
**Created:** [May 7, 2023, 2:23pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-finetune-whisper-model%2F38986 "2023-05-07T14:23:48Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Abdelkareem](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/abdelkareem/32/30422_2.png) [@Abdelkareem](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FAbdelkareem)%5C
**Post date:** [May 7, 2023, 2:23pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-finetune-whisper-model%2F38986%2F1 "2023-05-07T14:23:48Z")

</div>

I have read this tutorial: [Fine-Tune Whisper For Multilingual ASR with 🤗 Transformers](https://huggingface.co/blog/fine-tune-whisper)  
and when i go to the step of  
`#We can apply the data preparation function to all of our training examples using dataset's .map method common_voice = common_voice.map(prepare_dataset,remove_columns=common_voice.column_names["train"],num_proc=4)`  
It really take a loonnnnnng time and i must do it every run after shutdown my pc …etc is there any way to make easier not to train them once ! I don’t know how i am new to Transformers

- how can i fine tune the AJX version ?
- I tried to fine tune the model in kaggle but get string error : raise RuntimeError(f"Failed to load audio from {fileobj}")  
RuntimeError: Failed to load audio from \<\_io.BytesIO object at 0x77788b6c0900\>
- I triedy also to fine tuen it on colab but it get stuck in the step of common\_voice.map(…etc) and take a long time without any progress