I just updated my ‘Generate Alt Text with OpenAI Vision’ shortcut to version 1.1:
- Added an option to show the generated description at the end — a “test mode”, if you will.
- Added a step to convert .heic images to .jpeg before uploading them since the Vision API doesn’t support those.
Get it here from the HeyDingus Shortcuts Library.
Big thanks to @johnvoorhees for mentioning it in issue #400(!) of the @club newsletter. 🧡
it’s a wonderful shortcut, Jarrod, thank you so much. I have already integrated it in my other shortcuts and workflows and I’m really grateful. I had to tweak it a little, though, it was not sending the proper image to GPT, so I added this (sorry it’s in Spanish):
<img src=“https://umerez.eu/uploads/2024/img-0205.jpeg” width=“450” alt=“A smartphone screen displays an options menu with “Copiar Imagen convertida en el portapapeles” highlighted.
Una pantalla de smartphone muestra un menú de opciones con “Copiar Imagen convertida en el portapapeles” resaltado.”>
@eumrz I mean, I added that in the “media uploader” shortcut. And I used it to upload the former image, so I can’t believe how cool it is.
@eumrz Thank you for the kind words! Strange, it’s been sending the right images for me. I made a slight tweak that might fix it for the next version, but sounds like your solution is working well too!