modelscope/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR.git synced 2026-10-12 02:19:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat: add 'words' key aligned with timestamps in sensevoice model output (#2531)

Browse Source
This commit is contained in:
chengligen
2025-05-26 14:11:33 +08:00
committed by GitHub
parent 9038340be7
commit 8b0fb74bde
1 changed files with 8 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
funasr/models/sense_voice/model.py
+8 -3
View File
@@ -952,8 +952,8 @@ class SenseVoiceSmall(nn.Module):
timestamp.append([tokens[token_id], ts_left, ts_right])
token_id += 1
_start = _end
timestamp = self.post(timestamp)
result_i = {"key": key[i], "text": text, "timestamp": timestamp}
timestamp, words = self.post(timestamp)
result_i = {"key": key[i], "text": text, "timestamp": timestamp, "words": words}
results.append(result_i)
else:
result_i = {"key": key[i], "text": text}
@@ -962,6 +962,7 @@ class SenseVoiceSmall(nn.Module):
def post(self, timestamp):
timestamp_new = []
words_new = []
prev_word = None
for i, t in enumerate(timestamp):
word, start, end = t
@@ -972,17 +973,21 @@ class SenseVoiceSmall(nn.Module):
if i == 0:
# timestamp_new.append([word, start, end])
timestamp_new.append([start, end])
words_new.append(word)
elif word.startswith("▁"):
word = word[1:]
timestamp_new.append([start, end])
words_new.append(word)
elif prev_word is not None and prev_word.isalpha() and prev_word.isascii() and word.isalpha() and word.isascii():
prev_word += word
timestamp_new[-1][1] = end
words_new[-1] = prev_word
else:
# timestamp_new[-1][0] += word
timestamp_new.append([start, end])
words_new.append(word)
prev_word = word
return timestamp_new
return timestamp_new, words_new
def export(self, **kwargs):
from .export_meta import export_rebuild_model