heimoshuiyu/faster-whisper
1
0
Fork 0
Code Issues Pull Requests 1 Packages Projects Releases Wiki Activity

Check that the language code is valid

Browse Source
This commit is contained in:
Guillaume Klein
2023-02-21 12:10:54 +01:00
parent a98a2eeec4
commit f5c9f15c2c
1 changed files with 5 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
faster_whisper/transcribe.py
View File

@@ -107,8 +107,9 @@ class WhisperModel:
Arguments: Arguments:
input_file: Path to the input file or a file-like object. input_file: Path to the input file or a file-like object.
language: The language spoken in the audio. If not set, the language will be language: The language spoken in the audio. It should be a language code such
detected in the first 30 seconds of audio. as "en" or "fr". If not set, the language will be detected in the first 30 seconds
of audio.
task: Task to execute (transcribe or translate). task: Task to execute (transcribe or translate).
beam_size: Beam size to use for decoding. beam_size: Beam size to use for decoding.
best_of: Number of candidates when sampling with non-zero temperature. best_of: Number of candidates when sampling with non-zero temperature.
@@ -152,6 +153,8 @@ class WhisperModel:
language_token, language_probability = results[0][0] language_token, language_probability = results[0][0]
language = language_token[2:-2] language = language_token[2:-2]
else: else:
if self.tokenizer.token_to_id("<|%s|>" % language) is None:
raise ValueError("%s is not a valid language code" % language)
language_probability = 1 language_probability = 1
options = TranscriptionOptions( options = TranscriptionOptions(