Spoken language differs significantly from written language—this is a fact we must accept and always keep in mind when using a voice recorder for further work with the text we have dictated.
What is “forgiven” in conversation causes bewilderment in writing, and vice versa: terms, verbal techniques, and metaphors used in written language are often unacceptable in speech.
I decided to record my first article about why ComPost was created. To keep things fair, I created a special prompt—Author’s Blog Post—which essentially turns speech into the article itself.
I tried to make the prompt as universal as possible and suitable for any user’s goals. If writing articles is among your interests, you can easily download this prompt through the ComPost app, or go to this prompt right now, add it to your dashboard, and edit it to fit your needs.
After saving, the prompt will appear in your app in its edited form.
What can you do? You can add a website address, for example, https://mysite.com, and ask the AI to insert a link to this resource in the text output. You can clarify names and titles if you use them often and the AI constantly confuses them. For example, you can use a structure like this:
if I mention John Doe, link to his LinkedIn profile https://www.linkedin.com/in/johndoe/, but use this link only once for the entire text
In this case, the prompt (AI) will more accurately meet your exact requirements.
What is the result? Ultimately, publishing an article takes 5 minutes to record the audio + 30 seconds for the prompt to work + one minute for review and copying + 30 seconds to publish the article on a WordPress site (in my case). A total of 7 minutes where it used to take about an hour.
Try it out! ComPost saves time from the very first prompt.

Discussion
Discuss this article with other readers.