Figure out what we want to test

[waldoj]

• [ ] can retrieve video from legislature
• [ ] can save a copy of the video to the EC2 instance
• [ ] screenshots are extracted accurately (number, md5)
• [ ] the chyrons are being cropped properly (not too big, not too small)
• [ ] OCR gets chyrons within X% accuracy, for text and timestamp
• [ ] closed-caption text can be found in JSON format
• [ ] closed-caption text is correctly converted to SRT
• [ ] closed-caption text saves to the database correctly
• [ ] SRT is correctly converted to a transcript
• [ ] can save video to S3 bucket

[waldoj]

This will require creating a test video—something short enough to test quickly, but with enough chyrons to test meaningfully.

Also, this will require test caption JSON.

[waldoj]

OK, I extracted a 16-minute House video clip from the 2017 session that I think will work nicely. Now to figure out where to keep it. Maybe on the video S3 bucket?

[waldoj]

https://s3.amazonaws.com/video.richmondsunlight.com/test.mp4