Creating engaging e-learning videos with human presenters can be valuable, but it’s also time-consuming. Scheduling recordings, ensuring good audio quality, and finding the perfect background all have their challenges.
We’ve been updating our e-learning course on generative AI, and we were intrigued by the avatar feature in Audiate and wondered if it could streamline our video production.
Audiate, from TechSmith (the makers of Camtasia), is a text-based audio editing powerhouse. It transcribes audio into text, allowing you to edit the audio simply by editing the text. Beyond that, it also offers some interesting features like script generation, text-to-speech, and lip-synched avatars – which is what caught our eye.
The allure of avatars: efficiency and consistency
The appeal of using avatars for e-learning is clear:
- On-demand presenters: You can generate video content whenever you need it.
- Consistent backgrounds and look: Avatars ensure a uniform visual experience across your course, eliminating background distractions and maintaining a professional aesthetic.
- Potentially streamlined audio: While we used our own voice, Audiate also offers text-to-speech, potentially removing the need for voice recording altogether in some scenarios.
Audiate provides a diverse range of avatars, representing different ages, ethnicities, and genders, though currently, they all speak with an American accent.
We decided to put the avatar feature to the test.
Our experiment: audio recording + avatar sync
We opted to record our presentation audio with a human voice, and then synchronise it with an Audiate avatar. We initially tried processing a 30-minute recording, but we quickly learned that Audiate seemed to struggle with longer files. To overcome this limitation, we broke our recording into 3-minute segments. This proved to be a much smoother workflow within Audiate, allowing us to edit and synchronise each segment effectively.
Initial thoughts: promising, but with considerations
This initial experiment with Audiate avatars has been insightful. We can see potential for:
- Shorter video segments: Audiate seems to be most efficient with shorter audio clips, making it worth considering for micro-learning modules, quick updates, or supplementary course content.
- Projects with time or budget constraints: If you need to create video content quickly and cost-effectively, avatars offer an alternative to traditional video production.
- Internal training or rapid updates: For internal training materials or courses that require frequent updates, the speed and consistency of avatars could be a significant advantage.
However, it’s important to acknowledge the limitations:
- Expressiveness: While the avatars lip-sync well, they may lack the nuanced expressiveness and emotional connection of a real human presenter. This could be a factor depending on the subject matter and learning objectives.
- Potential “Uncanny Valley”: Some learners might still find them slightly unnatural compared to a human presenter. This is something to consider in terms of learner engagement.
Looking ahead
Our exploration of Audiate avatars is at the beginning.
This was only an experiment, and, for the moment, we intend to stick with human presenters for our e-learning course on generative AI.
We’ll continue experimenting to understand where they best fit within our course development workflow. We’re interested in testing different use cases and gathering learner feedback on avatar-presented content.
Have you experimented with avatars in e-learning? We’d love to hear about your experiences in the comments below.
Leave a Reply