Recently we started writing our product's user manual and just when we were nearing completion our users started asking for video walkthroughs of the product.
Needless to say, we started questioning ourselves whether we should have done the videos first. The outcome of that discussion is a topic for another day. Fow now, here's how we went about putting our product walkthrough videos together.
Before we started creating our walkthroughs we posed ourselves the following requirements.
Audio + Video - Keep text and annotations within the video to a minimum. We wanted the voice over to guide the user through the features. Use a text to speech service which will keep the voice over consistent across all our videos.
Ease of creation - Anyone from our (small) team should be able to create/edit these videos.
Maintainability - As we keep adding features to our product or change a functionality. We should be able to change the audio & video content with ease.
Budgetary constraints - We did not have funds allocated to create video walkthroughs. And thus, needed to create this at little or no cost.
After a couple of days of searching for tools required to create our walkthroughs, we settled on the following tools. It would be great to hear from you if there are (better) alternate tools available to perform the same function.
Keeping in mind the requirements, we wanted to make sure the audio files are easy to recreate in case of product updates. To create our audio content we settled on Amazon Polly. Amazon Polly is a service that turns text into lifelike speech.
Here is why we went with Amazon Polly:
We were able to create our audio content in Speech Synthesis Markup Language (SSML). Among other things, SSML makes it easy to add a pause in the audio to allow the screen to load or an action to be performed. It also makes the content easy to create / edit and add it to our source control. Our SSML files are now now part of our code documentation.
Under the free tier we got 5 million characters per month. This was more than we needed.
It has plenty of language and accent options. We went with the "Indian English" option as our target base at the moment is primarily Indian.
Using Amazon Polly we created .mp3 audio files for our voice over. Each audio file was kept under 2 minutes. This was done to make it easy to edit a particular section without impacting the full voice over.
Next we needed a screen capture tool which highlights the mouse pointer. For this, we used CamStudio. CamStudio is an open source desktop screen recorder. We would listen to the created audio and record our screen using CamStudio. This helped in keeping the video in sync with the voice over as far as possible.
To put this audio and video together we used OpenShot. OpenShot is an open source video editing software. Using OpenShot we were able to merge the recorded video and audio and create the final product.
Both the tools mentioned above have a lot of help content available online and are fairly easy to use.
Here is what our output looks like
It would be great to hear your feedback on this process and if you have used other tools which can make this easier.