Mirroristas/whisper
2
0
Fork 0
mirror of https://github.com/openai/whisper.git synced 2025-11-24 06:26:03 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge d5328a849f4e599706e33efdbbca37cff875a649 into c0d2f624c09dc18e709e37c2ad90c039a4eb72a2

Browse Source
This commit is contained in:
Gaspard Petit 2025-06-26 14:06:02 +08:00 committed by GitHub
commit c97d835dab
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
whisper/decoding.py
View File

@ -18,18 +18,20 @@ if TYPE_CHECKING:
@torch.no_grad()
def detect_language(
model: "Whisper", mel: Tensor, tokenizer: Tokenizer = None
) -> Tuple[Tensor, List[dict]]:
) -> Tuple[Tensor, Union[Dict[str,float], List[Dict[str,float]]]]:
"""
Detect the spoken language in the audio, and return them as list of strings, along with the ids
Detect the spoken language in the audio, and return them as a list of strings, along with the ids
of the most probable language tokens and the probability distribution over all language tokens.
This is performed outside the main decode loop in order to not interfere with kv-caching.
Returns
-------
language_tokens : Tensor, shape = (n_audio,)
ids of the most probable language tokens, which appears after the startoftranscript token.
language_probs : List[Dict[str, float]], length = n_audio
list of dictionaries containing the probability distribution over all languages.
ids of the most probable language tokens, which appear after the startoftranscript token.
language_probs : Union[Dict[str, float], List[Dict[str, float]]]
If the input contains a single audio sample, this will be a dictionary containing the
probability distribution over all languages for that sample. If the input contains multiple
audio samples, this will be a list of such dictionaries, one for each sample (length = n_audio).
"""
if tokenizer is None:
tokenizer = get_tokenizer(