Hot take: I think all these dictation tools are solving the wrong problem: they're optimizing for accurate transcription (and latency) when users actually need intelligent interpretation. For example: People don't speak in perfect emails. They speak in scattered thoughts and intentions that require contextual understanding.
Show HN: Whispering – Open-source, local-first dictation you can trust
141–150 of 159 posts
Re: Show HN: Whispering – Open-source, local-first dictation you can trust
#142I recommend adding support for nemo parakeet.
It's uncanny how good / fast it is
Re: Show HN: Whispering – Open-source, local-first dictation you can trust
#143Hot take: I think all these dictation tools are solving the wrong problem: they're optimizing for accurate transcription (and latency) when users actually need intelligent interpretation. For example: People don't speak in perfect emails. They speak in scattered thoughts and intentions that require contextual understanding.
Doesn't an accurate transcription make it easier to reach understanding?
Re: Show HN: Whispering – Open-source, local-first dictation you can trust
#144Among all the offline transcription apps I've tried, my favorite remains https://whispernotes.app . High accuracy, one-time purchase, and genuinely offline. I love its clean UI. Honestly, I'm getting tired of subscription-based apps. If it's truly offline, shouldn't it support a one-time purchase model? The whole point of local-first is that you're not dependent on ongoing cloud services, so why structure pricing lik…
Do you not see the case for Patreon at all? IAP subscriptions can be a form of Patreon for ongoing maintenance and R&D, even for offline applications.
I understand the fatigue but not the outright indignation.
Re: Show HN: Whispering – Open-source, local-first dictation you can trust
#145I'd love to find a tool which could recognise a few different speakers so that I could automatically dictate 1:1 sessions. In addition, I definitively would want to feed that to an LLM to cleanup the notes (to remove all "umm" and similar nonsense) and to do context aware spell checking. The LLM part should be very much doable, but I'm not sure if speaker recognition exists in a sufficiently working state?
Shameless plug -- check out speechischeap.com I spent three months perfecting the speaker diarization pipeline and I think you'll be quite pleased with the results.
How well does it work with multiple languages?
Re: Show HN: Whispering – Open-source, local-first dictation you can trust
#146Is there speaker detection?
Diarization is on the roadmap! Some providers support it, but some don't and the adapter for that could be tricky. Currently, for diarization I use the Elevenlabs Scribe API https://elevenlabs.io/app/speech-to-text , but there are surely other options
Do you know if there's any kind of writing about the different types of diarization methods?
Re: Show HN: Whispering – Open-source, local-first dictation you can trust
#147It already exists with great execution :
https://github.com/kitlangton/Hex
It translates to proper language also
Re: Show HN: Whispering – Open-source, local-first dictation you can trust
#148Are there any speech-to-text models that are fully OSS for everything from training data/code to model weights? https://salsa.debian.org/deeplearning-team/ml-policy
Not that I know of. I think the two most prominent open-source models that we hear about are Whisper and Parakeet!
Whisper doesn't list its training data (or code?), so can't be an open-source model, just an open weights model.
Parakeet does list its training data, and at least one of those is not FOSS, but some of them definitely are FOSS. I wonder if they nVidia would create a fully FOSS model by retraining on only the open data.
https://huggingface.co/nvidia/parakeet-rnnt-1.1b#datasets https://catalog.ldc.upenn.edu/LDC2004T19 https://catalog.ldc.upenn.edu/license/ldc-non-members-agreem...
Re: Show HN: Whispering – Open-source, local-first dictation you can trust
#149Reposting here, maybe you missed it where I asked first:
You mentioned that you got into YC .. what is the road to profitability for your project(s) if everything is open source and local?
Re: Show HN: Whispering – Open-source, local-first dictation you can trust
#150I would be very interested in a version of this that allow recording from both microphone and audio at the same time. Then it could get plugged into WhisperX for diarization..
But even just having anything that's being said recorded would be outstanding