Word Timestamp for whisper

Hi is there no way to get word timestamp using the whisper in candle?

The example successfully demonstrates the retrieval of segment timestamp but how would one retrieve word timestamp.

When I look into python code, they seem to pass this `word_timestamp=True` argument while transcribing and get the result with `base` model.

Is there any work around or can someone point me towards how to achieve this please.

Begin with the Candle Whisper example and compare its segment-timestamp retrieval with the Python transcription call using word_timestamp=True and the base model. Trace the transcription entry point and define done as a documented or working way to obtain word-level timestamps.

Written by the indexing model from the issue text.