Build an AI voice to text app w/ Bubble and the Whisper API in 20 minutes (no-code)

[@Tahycoon]

I watched 8 videos before yours and they just confused me, you're the best!!!

[@nocodemba]

Glad it helped!

[@superAweber]

No-code can be extremely complicated. Wow. An eye opening video for me.

[@AndreaPierros-fx4xp]

Thanks man, this was great! I’m working on building an app that provides real-time speech-to-text transcriptions for Zoom or Google Meet meetings. From a UX perspective, the app will allow users to see live transcriptions during their video meetings on my Bubble app. Can you help me understand how to implement this? Are there any resources you recommend checking out?

[@fin101-z3o]

Nice tutorial. Can i connect my own custom knowledge base so that responses have some context?

[@nocodemba]

Yup that's possible!

[@fin101-z3o]

How can I reach you

[@uen1857]

thanks a lot, i followed all steps as you and its working but it trasnlated the ausio to my native language even i was speaking in english, please how to prevent it from translation and it just transcribe it in the spoken language? is there any parameter to add to prevent tthe translation

[@OngoVonOTONGO]

Very well done. It appears to work overall, but I do not get the text output into the text field after I've spoken my text in there. Can you help?

[@OngoVonOTONGO]

Fixed it myself. Forgot to put the check on "save to bubble". (timestamp 14:45)

[@uen1857]

hello! it worked for me but it translated the audio to my native language, do you have any idea how to prevent it from translation and it transcribe it in the spoken language?

[@OngoVonOTONGO]

@@uen1857 you need to define the inpute language with another tool, e.g. google translate and use the language stamp e.g. in englisch "EN" in the prompt as a dynamic value.

[@uen1857]

@@OngoVonOTONGO thank you

[@Tahycoon]

Is there a way to auto-pause recording when the user stops talking just like with Siri and Alexa?

[@nocodemba]

Probably, but I haven't built it yet!

[@mixologymike]

Hi I set everything correctly( i think! but i'm getting a bug error code. I am not subscribed to Open AI or bubble yet.. is this why or did I just misconfigure something?

[@mixologymike]

Nevermind.. my bad.. I fixed it... awesome! thanks!

[@dhruvbahri9942]

i am getting the same thing API Status code 429, did u do anything else?

[@bluetensee]

Hi. thank you. Nice tutorial! Two questions: is using whisper for free if you're a plus subscriber (GPT4). And will this work on smartphone and with mp3 or youtube videos?

[@nocodemba]

Thanks! Whisper is a separate charge from being a plus subscriber. You pay for usage. Yeah it should work on a smartphone and with mp3 files! For youtube videos you'd just need to extract the audio.

[@bluetensee]

@@nocodemba super:) thank you so much!! :)

[@ki001]

@@bluetensee were you able to extract the audio and if so how?

[@bluetensee]

@@ki001 hi. No. Not really... Still waiting for an easy open source App to do that.

[@bryanstomp8578]

Hey man, nice video! The only thing I have trouble with is that I can only do one recording and it does not pick up the second recording. How can i fix this?

[@nocodemba]

Hey! What do you mean by a second recording?

[@bryanstomp8578]

@@nocodemba Well if i have recorded the first sentence, somehow it does not reset which causes that i am not able do more then one recording.

[@keysncoffee7733]

Hey brother. If you still need this: At 14:25 for the trigger "Do when AudioRecorder..." set the Run This dropdown to "Every Time" instead of "Just once" as shown here :) @nocodemba FYI -- I believe this is what he meant. Thanks for the great tutorial btw!

[@nocodemba]

@@keysncoffee7733 yes that should solve it, thanks for replying!

[@zenblue-bk]

So my API only work if you have chat gpt 4 ?

[@nocodemba]

Nope, it works as long as you have an account on OpenAI's platform

[@ki001]

Does not work on IOS