smshine/mediaplayer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fixed a bug

Browse Source
This commit is contained in:
Shuming Liu
2026-09-11 12:21:02 +08:00
parent 4b71c6c615
commit 67c7ad3d0f
1 changed files with 10 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
sentence_api/transcription.py
View File

@@ -99,7 +99,8 @@ class MossTranscriber:
segments = _parse_json_segments(payload.get("segments")) segments = _parse_json_segments(payload.get("segments"))
if not segments: if not segments:
segments = _parse_compact_segments(raw_text) segments = _parse_compact_segments(raw_text)
plain_text = " ".join(segment.text for segment in segments).strip() or raw_text plain_text = " ".join(segment.text for segment in segments).strip()
plain_text = _strip_speaker_tags(plain_text) or _strip_speaker_tags(raw_text)
return Transcript(text=plain_text, segments=segments) return Transcript(text=plain_text, segments=segments)
def _transcribe_chunked( def _transcribe_chunked(
@@ -142,7 +143,9 @@ class MossTranscriber:
) )
) )
segments.sort(key=lambda item: (item.start_seconds, item.end_seconds)) segments.sort(key=lambda item: (item.start_seconds, item.end_seconds))
plain_text = " ".join(segment.text for segment in segments).strip() plain_text = _strip_speaker_tags(
" ".join(segment.text for segment in segments).strip()
)
return Transcript(text=plain_text, segments=segments) return Transcript(text=plain_text, segments=segments)
def _post_audio( def _post_audio(
@@ -249,6 +252,11 @@ _ABBREVIATION_RE = re.compile(
_SPEAKER_TAG_RE = re.compile(r"\[(S\d+)\]") _SPEAKER_TAG_RE = re.compile(r"\[(S\d+)\]")
def _strip_speaker_tags(text: str) -> str:
"""Remove inline ``[S01]`` speaker tags from a text string."""
return _SPEAKER_TAG_RE.sub("", text).strip()
def _expand_speaker_segments( def _expand_speaker_segments(
segment: TranscriptionSegment, segment: TranscriptionSegment,
) -> List[TranscriptionSegment]: ) -> List[TranscriptionSegment]: