{"id":5045,"date":"2016-09-12T01:10:31","date_gmt":"2016-09-12T01:10:31","guid":{"rendered":"http:\/\/www.garysieling.com\/blog\/?p=5045"},"modified":"2016-09-12T01:10:31","modified_gmt":"2016-09-12T01:10:31","slug":"scoring-documents-quality-python-often-speaker-say-um","status":"publish","type":"post","link":"https:\/\/www.garysieling.com\/blog\/scoring-documents-quality-python-often-speaker-say-um\/","title":{"rendered":"Scoring documents for quality in Python – how often does a speaker say “um”?"},"content":{"rendered":"

As part of a project<\/a>, I thought it might be interesting to score lectures for how often the speaker says “um” (or similar). <\/p>\n

An interesting realization here is that an automated transcription of a lecture is superior for this purpose than manual closed captions or a written transcript, as those edit the material down.<\/p>\n

You need to tokenize whatever text you have:<\/p>\n

\nfrom nltk import word_tokenize\ntokens = word_tokenize(transcript)\n<\/pre>\n
Realistically, you only care if this is a frequent occurrence, so the best way to use this is combined with a threshold, or to feed this into a polynomial function that reduces the quality score for a transcript as it gets more severe.<\/p>\n
\ncheck = [\"um\", \"uh\", \"ah\", \"ehm\", \"eh\", \"uhm\", \"ah\", \"umm\", \"er\"]\n  \ndef umsScore(tokens):\n  bad = 0\n  for t in tokens:\n    if (t.lower() in check):\n      cnt = cnt + 1\n\n  return cnt\n<\/pre>\n","protected":false},"excerpt":{"rendered":"
Scoring talk transcripts for quality of word use in python<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"categories":[12],"tags":[447],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t