[Whisper] Fix decoder ids methods

[sanchit-gandhi]

What does this PR do?

The previous PR #20589 incorrectly returned a list of forced decoder ids:

from transformers import WhisperProcessor

processor = WhisperProcessor.from_pretrained("openai/whisper-tiny.en")
print(processor.get_decoder_prompt_ids(task="transcribe"))

Print Output:

[50257, 50358, 50362]

The correct format is a nested list of decoder ids, where the first element of each list specifies the position of the forced token and the second the token id:

print(processor.get_decoder_prompt_ids(task="transcribe"))

Print Output:

[(1, 50257), (2, 50358), (3, 50362)]

(at position 1 we force token 50257, at 2 we force 50358, at 3 we force 50362)

The PR also implements a test, thus making sure that no such error can be made again 😅

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

[HuggingFaceDocBuilderDev]

The documentation is not available anymore as the PR was closed or merged.

[sgugger]

Thanks for fixing!

[ArthurZucker]

Thanks for the quick fix

[ArthurZucker]

Nice catch! I should have realized when reviewing!

[ArthurZucker]

Nice!