zlacker

>>eloran+(OP)
Hi there! I'm the owner and operator of Broadcastify, which is the platform that powers all the apps that provide police scanners and public safety communications online. I'm an active HN reader and would be glad to answer any questions folks have.

It's an interesting business to be in these days...

>>blanto+h7
Is there a text transcript feature for users who may want to search through the communications? I'm curious how well those speech-to-text tools work for the audio feeds.

>>autojo+ww1
Hi, this is a difficult problem but I've been working hard on it for a couple of days with some help. I have a pipeline and website that automatically transcribes scanner feeds that is working pretty well, and the website allows users to correct and vote on transcriptions.

My goal is to train my own models on the corrected transcriptions (I work in the speech recognition space) so I can transcribe many live feeds inexpensively.

I will respond with a link here (hopefully very soon today) once I've fixed a couple of remaining UX bugs.

>>lunixb+rA1
I thought there were some open source speech-to-text models already [1].

Maybe there's something unique about how these low-quality radio transmissions sound that make these ineffective?

[1] https://voice.mozilla.org/en

>>ciaran+1B1
I work in the speech recognition space and train my own models already. The existing open-source models aren't very good at noisy radio speech. I will specialize one of my models to this task once I have some data from the site.