RFW0106: Report generator for STT_MV Saymore annotations

[spsither]

RFW0106: Report generator for STT_MV Saymore annotations

Named Concepts

• STT_MV: This refers to the Tibetan Movies department in Speech To Text.
• SayMore: A desktop application for Windows used for transcription.
• saymore-report-generator: The python script we use to generate the report currently.

Summary

Set up an STT_MV Saymore annotations report generator on Gade's PC.

Context

At the end of every month, the STT_MV department needs to come up with a report to calculate how much to pay based on the following factors for each STT_MV repository.

• time duration without speech
• syllables transcribed
• time duration with speech

Currently, we use saymore-report-generator but we need to set up the script so that Gade can do the same job without the need for a programmer to run the script.

Input

A list of STT_MV repository names

Expected Output

A CSV with the following columns

• Repo: Name of the STT_MV repository name. eg. STT_MV0790
• Blank segment time: Sum of time duration where the segment has no transcriptions
• Syllable count: Count of syllables in that Repo.
• With speech segment time: Sum of time duration where the segment has transcriptions
• Cost blank segment: Blank segment time * 5
• Cost syllable annotated: With speech segment time * 67
• Total cost: Cost blank segment + Cost syllable annotated

Expected Timeline

Jan 4th - 5th

References

• SayMore
• Elan

[kaldan007]

we need to specify the need of web app ccoz currently the script to generate repo is run mannually by developer. another solution would be prepare a jupyter note book for the annotator manager to run the report generator on his side.

https://github.com/OpenPecha/saymore-report-generator

[spsither]

Meet with Gade.