Workflow to generate text for captions

Currently, TechSmith (the people who produce Camtasia) have a tool called Audiate. They make it available as a fully functioning demo for 7 days (https://www.techsmith.com/audiate.html Links to an external site.) or on a monthly subscription. I found that the following workflow works well:

export your audio from Camtasia as a WAV file (Share->Export Audio Only)
Then enter the name for the WAV file that you are going to write
import this audio file into Audiate, then have it do the recognition. The time to do this is about the same time as the playtime for the audio clip. Note that there is a limit of 1 hour and 42 minutes for the duration of the audio file, if your file is longer than this you will need to split it up and feed it to Audiate in batches. As of mid-June in 2020, there is no warning about this, it will simply fail to complete.
This will give you a version of the file as recognized text and waveforms. Do not edit it in Audiate, as this will destroy the time synchronization with the video. Instead simply export the SRT file (this is a text file with lines of text in the form:
```
1
00:00:00,250 --> 00:00:02,590
Welcome to this module on professionalism and

2
00:00:02,590 --> 00:00:04,920
ethics for ICT students. My

3
00:00:04,920 --> 00:00:07,180
name is Gerald Maguire and I developed this

...
```
The next step is to edit this SRT file (I generally use emacs for this) while watching and listening to your presentation- At this point, I do two things:
1. insert text of the form: [slide 2] into the caption text at the point when slide 2 is shown
2. fill in missing or misrecognized text for the caption
Note that to help listen to the audio - you can use the Kaltura short cuts - see Keyboard shortcuts when watching videos.
In the next step, I take the resulting SRT file and either edit into a text transcript (i.e., removing the lines with the caption number, the start and end times, and the blank line following the caption) and insert this into Word or instead of editing directly, I run a program to convert the SRT file into an XLSX spreadsheet (see srt2spreadsheet.py Links to an external site. at https://github.com/gqmaguirejr/Camtasia_and_Audiate_tools Links to an external site.) and then select the column with all the captions and copy&paste this into Word.
In Word, I convert the text of the form "[slide 2]" into a heading and then leave the rest of the captions in this "section".
Next, I use Grammarly to check through the text. When I find an error, then I:
- correct the SRT file in my editor and
- make the correction in the Word file.
When finished with the previous step, you have a correct SRT file and a transcript of the presentation as a word file, for example:
```
Ethical-research-edited.srt
Ethical-research-transcript.docx
```
Now you can import the SRT file into the video (see next page).
I also created folders Captions and Transcripts in each Canvas course to be able to store the caption files and the transcripts (in DOCX format). Later, I add a link to the transcript for the video to the page where the video appears.

Alternative - If you are using TechSmith's Camtasia

Camtasia has a speech to text conversion function in it if you are on a Windows 10 machine, there is a tutorial at:
https://support.techsmith.com/hc/en-us/articles/203729278-Camtasia-Windows-How-to-Configure-Microsoft-Speech-Engine-for-Speech-To-Text?source=search Links to an external site.

Transcript

Degree_projects_hints_transcript.docx Download Degree_projects_hints_transcript.docx