Open jersonal-com opened 7 months ago
Hi @jersonal-com,
I have had in the back of my mind for a while, an option to save notes from episodes via the transcripts (for those podcasts with transcripts). I hadn't considered using speech to text - it's an interesting idea. There are a few things in the queue at the moment, but I'll try to find some time to take a look at this library.
My only concern with using speech to text is the real-time nature of generating the note, i.e. you have to play the audio clip in real time to generate the text before you can save it as a note. Using an existing transcript, the user would have the option of either directly selecting a block of text, or selecting a snippet of audio and saving that part of the transcript as a note.
Hi @amugofjava,
Thanks for the quick reply. The Momento app also seems to rely on speech to text (when it was working - it seem broken since some time) and at least for my usage pattern realtime is not important. My typical usage would be like so:
I don't know if my usage is typical but given the implementation in other apps it seems to be a common pattern that recording and editing are time-wise separated tasks.
Hi @jersonal-com,
Thanks for the further details - that workflow makes complete sense and is, I believe, how Snipd implements its note taking.
I like this idea and can see it being very useful, but there are several other items in the queue so it may be some time before I have an opportunity to look at this - unless someone in the community would like to pick this up.
Some podcasts are not just entertaining but also contain valuable insights. There are a few apps like snipd or Momento that offer the user to save snippets from the podcast as text.
However they are closed source, buggy and do not work well with other OSS solutions like LogSeq.
Given that there is a speech_to_text Module for flutter I was wondering if it would be feasible to integrate such a feature into the Anytime player?