bugfix/fix-newline-webvtt-sentence-segmentation

cdp_backend/sr_models/webvtt_sr_model.py

@@ -164,7 +164,15 @@ def _get_sentences(
         # List of text, representing a sentence
         lines: List[str] = []
         start_time: Optional[float] = None
+
         for caption in speaker_turn_captions:
+
+            # Clean text of line breaks
+            caption.text = caption.text.replace("\n", " ")
+
+            # Remove any double spaces as result of line break removal
+            caption.text = caption.text.replace("  ", " ")
+
             if start_time is None:
                 start_time = caption.start_in_seconds
             lines.append(caption.text)

cdp_backend/tests/resources/boston_captions.vtt

@@ -0,0 +1,36 @@
+WEBVTT
+
+00:00:00.000 --> 00:00:00.000
+
+
+00:00:40.337 --> 00:00:42.272
+uh uh,
+I'm the city councilor for
+
+00:00:44.975 --> 00:00:46.909
+district five and I'm the vice
+PRESIDENT Of the boston city
+council.
+
+00:00:47.578 --> 00:00:49.445
+Viewers can watch the
+council meeting live on youtube
+
+00:00:51.415 --> 00:00:53.783
+by visiting boston city
+council tv.
+I would like to ask my
+
+00:00:54.985 --> 00:00:57.653
+colleagues and those in the
+audience to please silence
+their phones and electronic
+devices.
+
+00:00:58.522 --> 00:01:00.590
+Thank you.
+Please also be respectful and
+do not disrupt the meeting
+
+00:01:00.590 --> 00:01:02.590
+while you are here.