- accept transcript CSV column names (speaker_uid, utterance_text, start_time, end_time, utterance_index) as utterance input
- parse string timestamps as floats
- add talkTimeDistribution, talkTimeLength, talkTimeTotal, talkTimeBySpeaker, sessionDuration and talkTimeUntimedUtterances (seconds)
- add duration (seconds) to each talkMoments item
- return (None, None) from get_talk_distribution_and_length when uptake_speaker is missing instead of crashing the handler
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- testing math vocabulary measure (272cb8dc30cc1a9f8f2b36e84c95f3e7312c9244)
- extend response with matches (43d54f4fd3b88c352a644836f5fa7c840957ab55)
- return matches to the client (be125188af76a41802957611aedc2a3cfde7a509)
