Skip to content

Latest commit

 

History

History
78 lines (48 loc) · 2.81 KB

NOTES.md

File metadata and controls

78 lines (48 loc) · 2.81 KB
Raw

Transcribe a Session

Determine the docLabel for the Session

Download Session Audio

Download M4A file from Youtube

Split Up Large Session Files (Symposium)

Upload M4A file to Arweave

Generate Transcript

Create the word_boost.txt file

Listen to the session and watched the scrolling transcript. Write down two kinds of phrases (one or more words):

  • Important ones
  • Ones that are misspelled in the scrolling CC (or seem likely to be misspelled)

don't use hyphens in the phrases

Mahault Albarracin
Active Inference
deontic value

save the file to the session's metadata folder: word_boost.txt

Create the custom_spelling.csv file

A single word (no spaces nor hyphens), which (I think) is treated as case insensitive inside AssemblyAI; a comma, then a "to" word with the correct spelling.

actinf,ActInf
bayesian,Bayesian

save the file to the session's metadata folder: custom_spelling.txt

save the file to the session's metadata folder: speakers.csv

Call AssemblyAI

fill in the API Key into the authkey.txt file

cd into the session's metadata folder

run this command:

python '/mnt/md0/projects/Journal-Utilities/2_audio_to_markdown/SubmitToCloudWhisper.py' 'CCL2023-08' 'wMeQ52fdz_U1vbhGE4mmiDYCFlc1QgH3CFrv3Ah2HXg' ONLINEPATH 'https://arweave.net' AUTHKEYFILENAME '/mnt/md0/projects/Journal-Utilities/2_audio_to_markdown/authkey.txt' WORD_BOOST_FILE_LIST 'word_boost.txt' SENTIMENT_ANALYSIS False IAB_CATEGORIES False CUSTOM_SPELL_BOOSTED True CUSTOM_SPELLING_FILE_LIST 'custom_spelling.csv' | tee 'trace.txt'

Create the Editable Transcript

Create an speakers.csv file

Look at <filename>.paragSpeakers.csv to create <filename>.speakers.csv in this format:

DocLabel,Speaker Label,Displayed Speaker Name,Full Speaker Name,First Turn,RangeFrom,RangeTo,Notes,,,,
CCL2023-08,A,Daniel,Daniel Friedman,00:08,,,,,,,
CCL2023-08,B,Mahault,Mahault Albarracin,00:29,,,,,,,

Save in the session's metadata folder

Run the sentencesToTranscripts.py script

cd to the session's Transcripts/Prose folder and run this command:

python '/mnt/md0/projects/Journal-Utilities/2_audio_to_markdown/sentencesToTranscripts.py' 'CCL2023-08' '/mnt/md0/projects/ActiveInferenceJournal/Courses/ActiveInferenceForTheSocialSciences/NormsScripts_Lecture/Metadata' 'CCL2023-08_wMeQ52fdz_U1vbhGE4mmiDYCFlc1QgH3CFrv3Ah2HXg.sentences.csv' INSPEAKERDIR '/mnt/md0/projects/ActiveInferenceJournal/Courses/ActiveInferenceForTheSocialSciences/NormsScripts_Lecture/Metadata' SPEAKERFILE 'CCL2023-08_wMeQ52fdz_U1vbhGE4mmiDYCFlc1QgH3CFrv3Ah2HXg.speakers.csv' | tee CCL2023-08.m4a_transcript.json

/mnt/md0/projects/ActiveInferenceJournal/Courses/ActiveInferenceForTheSocialSciences/NormsScripts_Lecture/Metadata/CCL2023-08_wMeQ52fdz_U1vbhGE4mmiDYCFlc1QgH3CFrv3Ah2HXg.speakers.csv