Fix occasional IndexError on empty segments (#227)

This commit is contained in:
Guillaume Klein
2023-05-24 12:55:04 +02:00
committed by GitHub
parent 6a2da9a95c
commit 723cb97483

View File

@@ -763,6 +763,8 @@ class WhisperModel:
text_tokens + [tokenizer.eot] text_tokens + [tokenizer.eot]
) )
word_boundaries = np.pad(np.cumsum([len(t) for t in word_tokens[:-1]]), (1, 0)) word_boundaries = np.pad(np.cumsum([len(t) for t in word_tokens[:-1]]), (1, 0))
if len(word_boundaries) <= 1:
return []
jumps = np.pad(np.diff(text_indices), (1, 0), constant_values=1).astype(bool) jumps = np.pad(np.diff(text_indices), (1, 0), constant_values=1).astype(bool)
jump_times = time_indices[jumps] / self.tokens_per_second jump_times = time_indices[jumps] / self.tokens_per_second