AI-Generated Transcriptions for UX Research: Where Are We?



29 November 2021


Speech to Text Icons

From self-driving cars to the Internet of Things, Artificial Intelligence is progressing at an impressive rate. While not still reaching the crazy human-like features foretold by 1950s Sci-Fi novelists, Artificial Intelligence has turned countless manual tasks into automated ones.

In UX research, AI-generated video and audio transcription above all has completely revolutionised how UX Researchers conduct qualitative researches, saving them countless hours of manual work.

Without the hassle of checking and transcribing tens of hours of interviews, UX specialists can now do larger and deeper studies, improve data quality, and reduce data collection time.

Speech to text Icon, light blue bg

However, that’s just one part of the story. With many critics arguing automated transcriptions lack accuracy, context and are of mixed quality, there is some room for debate.

Cutting to the chase — what are the concrete benefits of automated transcription, and how does it compare against its manual counterpart?

Faster, Cheaper, Easier: 3 benefits of automated transcription for UX’ers

AI-generated transcription is faster. Where manual transcription requires on average of 4x the duration of the video, AI can transcribe at the same speed of the recording — when not faster. Even accounting for proofing, this is massive in terms of time saved.

AI-generated transcription is cheaper. Indeed, where manual transcription services cost from 60$ per hour of audio or video on the low end, proprietary automated transcription tools are on the market for far less, and are in many cases bundled with other AI-powered tools that save even more time to UX Researchers!

laptop icon with light blue background

Finally, as the quality of AI in Automatic Speech Recognition has improved in leaps and bounds over the last years, AI can produce highly accurate transcripts, making automated transcription easier to adopt than in the past.

Get prime AI transcriptions with Sonar — and more!

So, are there no downsides to adopting automated transcriptions into your qualitative research? Well… No. While automated transcriptions are accurate, they are not as accurate as manual ones. Furthermore, AI may provide transcriptions that do not take into account the context or the complexity of the vocabulary being used in the interviews, resulting in constant mistakes.

How do you avoid this pitfall?

Though there are a number of very cheap audio and video transcriptions tools available, if you are using an AI fine-tuned with the lingo and context of your study, you can reliably increase accuracy, build a custom vocabulary and get so much more to boost your qualitative study!

Sonar’s own AI-powered transcription tool allows you to input custom words to improve the quality of your transcriptions. Furthermore, leveraging the knowledge built through hundreds of thousands of interviews, Sonar AI is one of the best in the market for UX Research and qualitative studies.

But that’s not the end of it! Our comprehensive platform allows designing, collecting, analysing and sharing user research faster and more easily.

From automatic quote generation to intuitive report builders, you can truly delve into deep analyses and leave every boring bit of your research to our platform.


Curious to see other powerful features to boost your research? Just click on the button below!

Get the latest news from Sonar


About Us
Cookie Policy
Privacy Policy


Langebrogade 4
1411 Copenhagen

Phone: +45 7734 8685
VAT: 33510608




Get in touch

Book a demo
Write to us