You can use Dragon© Naturally Speaking to transcribe your videos*. You can also use a protocol developed by Engineering-LAS Online Learning (ELO) to caption existing videos. This protocol requires Dragon Naturally Speaking, a speech-to-text software. The protocol currently works only on the Windows OS version of Dragon Naturally Speaking.
Tests done by ELO faculty and Office of Digital Access showed that following protocol shortens the amount of time allocated for transcription and captioning of the videos.
Protocol of Transcription and Captioning with Dragon© Naturally Speaking (Windows OS Version)
Once you have a recorded video in MP4 format,
- Create and customize your Dragon Naturally Speaking profile by training it. Select and utilize one of the available training texts provided by the software. Choosing a text that matches the complexity of your video topics will yield increasingly efficient results for the transcription process.
- Separate the audio from your video by using VLC Media Player or Handbreak. Save your audio as MP3 file (Dragon will only accept MP3 files). Both of these resources are open source and freely available for you to download.
- Upload the MP3 file to Dragon Naturally Speaking to transcribe. This process can take some time depending on the length of your audio file.
- Once you have the transcript, correct it.
- Re-train dragon with the corrected version of the transcript to increase the accuracy in transcription for future transcriptions. Don't skip this step if you want to increase accuracy and decrease processing time in future transcriptions.
- Upload the video to YouTube along with the transcript to create the time stamps for the transcription and close captioning your video.
*This protocol is meant as a suggestion and it does not constitute a product endorsement.