def compose_from_transcript(files, search, searchtype): """Takes transcripts created by audiogrep/pocketsphinx, a search and search type and returns a list of timestamps for creating a supercut""" final_segments = [] if searchtype in ['re', 'word', 'franken', 'fragment']: if searchtype == 're': searchtype = 'sentence' segments = audiogrep.search(search, files, mode=searchtype, regex=True) for seg in segments: seg['file'] = seg['file'].replace('.transcription.txt', '') seg['line'] = seg['words'] final_segments.append(seg) elif searchtype in ['hyper', 'pos']: for s in audiogrep.convert_timestamps(files): for w in s['words']: if search_line(w[0], search, searchtype): seg = { 'file': s['file'].replace('.transcription.txt',''), 'line': w[0], 'start': float(w[1]), 'end': float(w[2]) } final_segments.append(seg) return final_segments
def compose_from_transcript(files, search, searchtype): """Takes transcripts created by audiogrep/pocketsphinx, a search and search type and returns a list of timestamps for creating a supercut""" final_segments = [] if searchtype in ['re', 'word', 'franken', 'fragment']: if searchtype == 're': searchtype = 'sentence' segments = audiogrep.search(search, files, mode=searchtype, regex=True) for seg in segments: seg['file'] = seg['file'].replace('.transcription.txt', '') seg['line'] = seg['words'] final_segments.append(seg) elif searchtype in ['hyper', 'pos']: for s in audiogrep.convert_timestamps(files): for w in s['words']: if search_line(w[0], search, searchtype): seg = { 'file': s['file'].replace('.transcription.txt', ''), 'line': w[0], 'start': float(w[1]), 'end': float(w[2]) } final_segments.append(seg) return final_segments