Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Whispering – Open-source, local-first dictation you can trust

github.com

Post reply on HN
Print view View original

101–110 of 159 posts

Re: Show HN: Whispering – Open-source, local-first dictation you can trust

#101

Post by braden-w »

sa-code wrote: post #52
satvikpendem wrote: post #29

All these all just Whisper wrappers? I don't get it, the underlying model still isn't as good as paid custom models from companies, is there an actual open source / weights alternative to Whisper for speech to text? I know only of Parakeet.

Voxtral mini is a bit bigger but their mixed language demos looked super impressive https://mistral.ai/news/voxtral

We like Whisper because it's open-source :) but we also support OpenAI 4o-transcribe/ElevenLabs/Deepgram APIs that all use non-Whisper models (presumedly) under the hood. Speaches also supports other models that are not Whisper. Hopefully adding Parakeet support later too!

Re: Show HN: Whispering – Open-source, local-first dictation you can trust

#102

Post by braden-w »

tummler wrote: post #24

Related, just as a heads up. I've been using this for 100% local offline transcription for a while, works well: https://github.com/pluja/whishper

Awesome, thank you so much for bringing this to my attention and including it in the thread! Always cool to see other open source projects :)

Re: Show HN: Whispering – Open-source, local-first dictation you can trust

#103

Post by braden-w »

hephaes7us wrote: post #58

Thanks for sharing! Transcription suddenly became useful to me when LLMs started being able to generate somewhat useful code from natural language. (I don't think anybody wants to dictate code.) Now my workflow is similar to yours. I have mixed feelings about OS-integration. I'm currently working on a project to use a foot-pedal for push-to-transcribe - it speaks USB-HID so it works anywhere without software, and it…

Thank you for the support, and agreed on OS-level integration. At least for me, I have trouble trusting any app unless they are open source and have a transparent codebase for audit :)

Re: Show HN: Whispering – Open-source, local-first dictation you can trust

#104

Post by braden-w »

jryio wrote: post #93
nchudleigh wrote: post #86

Earlier quoted context omitted.

superwhisper has that functionality.

Right but not running locally on device. No privacy

I really want to run it locally on a phone, but as a developer it's scary to think about making a native mobile app and having to work with the iOS toolchain I don't have bandwidth at the moment, but if anyone knows of any OSS mobile alternatives, feel free to drop them!

Re: Show HN: Whispering – Open-source, local-first dictation you can trust

#105

Post by _345 »

braden-w wrote: post #99
hephaes7us wrote: post #77

Earlier quoted context omitted.

Have you tried Whisper itself? It's open-weights. One of the features of the project posted above is "transformations" that you can run on transcripts. They feed the text into an LLM to clean it up. If you're willing to pay for the tokens, I think you could not only remove filler-words, but could probably even get the semantically-aware editing (corrections) you're talking about.

^Yep, unfortunately, the best option right now seems to pipe the output into another LLM to do some cleanup, which we try to help you do in Whispering. Recent transcription models don't have very good built-in inference/cleanup, with Whisper having the very weak "prompt" parameter. It seems like this is probably by design to keep these models lean/specialized/performant in their task.

By try to help, do you mean that it currently does so or that functionality is otw

Re: Show HN: Whispering – Open-source, local-first dictation you can trust

#106

Post by braden-w »

0xbadcafebee wrote: post #51

Not a fan of high resource use or reliance on proprietary vendors/services. DeepSpeech/Vosk were pre-AI and still worked well on local devices, but they were a huge pain to set up and use. Anyone have better versions of those? Looks like one successor was Coqui STT, which then evolved into Coqui TTS which seems still maintained. Kaldi seems older but also still maintained. edit: nvm, this overview explains the differ…

Sorry for the delayed response, thank you for sharing these articles! I agree. I hope that we get a lot better open-source STT options in the future.

Re: Show HN: Whispering – Open-source, local-first dictation you can trust

#107

Post by dang »

Aachen wrote: post #47

Wait, I'm confused. The text here says all data remains on device and emphasises how much you can trust that, that you're obsessed with local-first software, etc. Clicking on the demo video, step one is... configuring access tokens for external services? Are the services shown at 0:21 (Groq, OpenAI, Antrophic, Google, ElevenLabs) doing the actual transcription, listening to everything I say, and is only the resulting…

We've edited the top text to make this clearer now. Thanks for pointing this out!

Re: Show HN: Whispering – Open-source, local-first dictation you can trust

#108

Post by braden-w »

hn1986 wrote: post #78

excellent tool and easy to get started. on win11, i installed ffmpeg using winget but it's not detecting it. running ffmpeg -version works but the app doesn't detect it. one thing, how can we reduce the number of notifications received? i like the system prompt option too.

Thank you for the support! Sorry for the issues with FFmpeg. This is an active issue that we're tracking:

https://github.com/epicenter-so/epicenter/issues/674

We hope to fix notifications too thank you for the feedback and happy to hear you liked the system prompt!

Re: Show HN: Whispering – Open-source, local-first dictation you can trust

#109

Post by teiferer »

braden-w wrote: post #4

For those checking out the repo this morning, I'm in the middle of a release that adds Whisper C++ support! https://github.com/epicenter-so/epicenter/pull/655 After this pushes, we'll have far more extensive local transcription support. Just fixing a few more small things :)

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

#110

Post by braden-w »

emacsen wrote: post #62
emacsen wrote: post #61

Tried it with AppImage on Linux, attempted to download a model and "Failed to download model. An error occurred." but nothing that helps me track down the error :(

Same with the deb. :(

Thanks for flagging this, and sorry that this is happening! Does downloading the model manually work? I wonder if it's related to this:

https://github.com/epicenter-so/epicenter/issues/669

Post reply on HN