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

Transcribed police scanners in real-time

murph.live

Post reply on HN
Print view View original

11–20 of 109 posts

Re: Transcribed police scanners in real-time

#11

Post by gfosco »

bberenberg wrote: post #4
cddotdotslash wrote: post #2

I really like this idea - it also has the side effect of making all of these streams easily searchable later on. That being said, transcription still isn't at the point where it can understand radio garble. > because I'm going to take that group that someone over the Willing. Can you raise Lieutenant lady and having semi-open Buddhist month > you know the carrot for you and Hudson no accident Reminds me of the early…

Absolutely agree. I feel like the next step here would be to: * Put the audio clip next to the transcribed text * Flag transcribed text which fails some heuristics * Allows users to suggest edits to the transcription * Train the transcription models on contributed edits + police specific lingo

Cooperative community editing, genius-style, would be perfect. Big events would lead to incredibly accurate transcripts.

Re: Transcribed police scanners in real-time

#12

Post by lunixbochs »

blantonl wrote: post #6

This is very impressive. I'm the owner of Broadcastify.com, where presumably these streams are being transcribed from. We've dabbled in this space and looked at real-world approaches to taking something like this to market, but transcribing 7000+ streams to text seems like an expensive (computational) and ($$) effort that needs a lot of investigation. Note to mention that the individual lexicons between streams are d…

I prototyped this concept too, at https://feeds.talonvoice.com with prohibitively expensive Google speech recognition, but also have a feature for users to listen and fix transcriptions. If murph was anything like me they probably paid for broadcastify and tailed a couple of the static mp3 feeds.

My plan was to collect user transcription corrections on my site then train my own inexpensive models on them. The open-source speech tech I work on can do passable transcription at close to 100x faster than realtime on a quad core desktop CPU (or 200 simultaneous streams per 4-core box at 50% activity). With higher quality transcription it's closer to 10-20x faster than realtime.

For your case you could also try to push some of the computation down to the uploading machine. These models can run on a raspberry pi.

I think the biggest work for a new effort here is going to be building local language models and collecting transcribed audio to train on. However, there have been a couple of incredible advances in the last year for semi-supervised speech recognition learning, where we can probably leverage your 1 year backlog as "unsupervised training data" while only having a small portion of it properly transcribed.

The current state-of-the-art paper uses around 100 hours of transcribed audio and 60,000 hours of unlabeled audio, and I bet you could push the 100h requirement down with a good language model and mixing in existing training data from non-radio sources.

Re: Transcribed police scanners in real-time

#13

Post by blantonl »

robbiet480 wrote: post #9
blantonl wrote: post #6

This is very impressive. I'm the owner of Broadcastify.com, where presumably these streams are being transcribed from. We've dabbled in this space and looked at real-world approaches to taking something like this to market, but transcribing 7000+ streams to text seems like an expensive (computational) and ($$) effort that needs a lot of investigation. Note to mention that the individual lexicons between streams are d…

Hey Lindsay, I'm the one who just added EBRCS to Calls. Wondering if a solution to both cost problems would be to (optionally) have submitters upload a transcript along with each call? Could build a model into trunk-recorder maybe?

That might be an option, but we'd have to somehow get trained models down to the client, which is one of the issues.

We're working on client ingest models now that work on more of a "tasking" perspective, where someone deploys a device that is GPS enabled and then we send an ingest task to fill in coverage, start new coverage, etc. But this is predicated on low cost ingest devices (read: RPi and RTL sticks) which might not have the horsepower needed for transcription at the client level.

Re: Transcribed police scanners in real-time

#14

Post by elicash »

rachelbythebay wrote: post #8

Example audio: https://www.broadcastify.com/webPlayer/29351 This is close enough to the Seattle feed that you can do a compare & contrast. Heard: "clear my first call ocean nora 470" On the site: "charlie my first call" So, yeah, this still has a long long way to go. I considered and discarded this in 2011 because it was pure insanity, and as another comment suggests, it's highly context-sensitive. "ECR" is El Camino…

Seems like there could be a lot of use cases where you don't need anything even close to perfection.

Re: Transcribed police scanners in real-time

#15

Post by wfbarks »

cddotdotslash wrote: post #2

I really like this idea - it also has the side effect of making all of these streams easily searchable later on. That being said, transcription still isn't at the point where it can understand radio garble. > because I'm going to take that group that someone over the Willing. Can you raise Lieutenant lady and having semi-open Buddhist month > you know the carrot for you and Hudson no accident Reminds me of the early…

"strength tenets we do not have any to 9 available yeah five-year-old advisor. We have several dogs around miracles"

Re: Transcribed police scanners in real-time

#17

Post by blantonl »

lunixbochs wrote: post #12
blantonl wrote: post #6

This is very impressive. I'm the owner of Broadcastify.com, where presumably these streams are being transcribed from. We've dabbled in this space and looked at real-world approaches to taking something like this to market, but transcribing 7000+ streams to text seems like an expensive (computational) and ($$) effort that needs a lot of investigation. Note to mention that the individual lexicons between streams are d…

I prototyped this concept too, at https://feeds.talonvoice.com with prohibitively expensive Google speech recognition, but also have a feature for users to listen and fix transcriptions. If murph was anything like me they probably paid for broadcastify and tailed a couple of the static mp3 feeds. My plan was to collect user transcription corrections on my site then train my own inexpensive models on them. The open-so…

Our new project, Broadcastify Calls, might be a better fit for this. Instead of 24x7 live streams, we capture and ingest every individual call as a compressed audio file from SDRs (software defined receivers) We can then ingest and present back to consumers playback, rewind, playlist, of those calls. We're now capturing over 100 systems and 800-900 calls a minute... as we solidify the architecture it will be our new direction for how we capture and disseminate public safety audio (Police Scanners)

https://www.broadcastify.com/calls

Re: Transcribed police scanners in real-time

#18

Post by lunixbochs »

pugworthy wrote: post #16

Some of the text though are pretty weak. I have no idea if this is just garbled transcription, or dreamy cop talk... "movie up and around pain and dreamy I'm not" or "happy person that you can just so may I help you"

It's the transcription, I think this kind of site should allow you to fall back to listening to the audio. Even with high accuracy, if there's no human curation, there will be pretty awkward mistakes every once in a while in the text.

Re: Transcribed police scanners in real-time

#19

Post by lunixbochs »

blantonl wrote: post #17
lunixbochs wrote: post #12

Earlier quoted context omitted.

I prototyped this concept too, at https://feeds.talonvoice.com with prohibitively expensive Google speech recognition, but also have a feature for users to listen and fix transcriptions. If murph was anything like me they probably paid for broadcastify and tailed a couple of the static mp3 feeds. My plan was to collect user transcription corrections on my site then train my own inexpensive models on them. The open-so…

Our new project, Broadcastify Calls, might be a better fit for this. Instead of 24x7 live streams, we capture and ingest every individual call as a compressed audio file from SDRs (software defined receivers) We can then ingest and present back to consumers playback, rewind, playlist, of those calls. We're now capturing over 100 systems and 800-900 calls a minute... as we solidify the architecture it will be our new…

The source repo to feeds.talonvoice.com includes a test ingestor that scrapes your calls API and uploads the src/dst info with the transcription, I haven't tested it live though.

Re: Transcribed police scanners in real-time

#20

Post by godzillabrennus »

blantonl wrote: post #6

This is very impressive. I'm the owner of Broadcastify.com, where presumably these streams are being transcribed from. We've dabbled in this space and looked at real-world approaches to taking something like this to market, but transcribing 7000+ streams to text seems like an expensive (computational) and ($$) effort that needs a lot of investigation. Note to mention that the individual lexicons between streams are d…

You could conceivably do this using the text to speech recognition on a Mac using Loopback to capture the stream to a microphone input.

https://rogueamoeba.com/loopback/

Someone clever enough could create containers to run the software locally and have many loops running off many streams to many instances of the audio to text feature.

Post reply on HN