The next update to Leasey version 10 is almost complete and it is expected that it will be available after the Easter break, around 22 April.
The feedback we have received suggests that many people are enjoying (and making great use of) the ChatGPT integration, particularly now it can pull information from the web to get real-time results. However, there is at least one area where ChatGPT does not perform particularly well, and that is in respect of audio transcription: transcribing an audio file to text.
For this reason, the April update moves away from ChatGPT for this purpose and will be using alternative functionality.
What does this mean for you?
More accurate transcription. The text recognition is significantly improved. You are able to select whether you would like Leasey to differentiate between different speakers (or narrators), or to just output the full text without separation.
You will be able to transcribe longer files. We highly recommend that a benchmark to use for audio length is two hours. The content is rendered in your web browser initially. Currently, JAWS is seriously struggling to work with lengthy pages, so please keep this very much in mind.
Full instructions will be provided as part of the Leasey documentation when the new update is available.
We hope you will benefit from this new improvement!