willwade

Are you not looking for ssml with ipa tag? I think you might be. It’s part of all your standard OS tts - including espeak-ng on Linux. Also in Google cloud, azure, Watson, and Amazon Polly voices.

Nice. Stick a licence on that and join https://www.openaac.org/

Please OP follow this.

There’s a LOT of similar apps in this space. But that doesn’t remove the interest - the backstory is key. And is it open in anyway? ~What voice engine are you using?~ (11labs I see - fall back to browser voice?). Why Make this and the other solitons didn’t meet this ?

Nb. Join openaac if it’s open :-)

anyone got a paywall workaround for this?

I built an error correction model for people with significant disabilities who can’t speak and poor finger control to write out their message. https://willwa.de/2024/01/12/rapid-development-in.html app: look for dragtospeak on AppStore and model https://huggingface.co/willwade/t5-small-spoken-typo

Like SSML? See azure tts or google cloud tts, or ibm Watson or even old school system tts like SAPI voices on windows. But I hear you. In a VITS typical model system ssml isn’t standard. Piper tts does have it on the roadmap.

- see hable one. A few FOSS variants have been tried but failed - look up “switch boxes” - look up “eyegaze” hardware

Truth is hardware is actually hard to get absolutely right. Software - now there’s tons of space on this. The problem is identifying the big pinch points. If anyone there wants to help us look at these priorities

- ows. An open wireless switch box https://github.com/AceCentre/OWS/tree/main (largely hardware) - echo. SwiftUI app. Aiming to provide speech to blind and physically impaired and with no voice. https://github.com/AceCentre/Echo - facecommander. A fork of googles gameface that was appallingly bug ridden. Use your face gestures to act as keyboard inputs https://github.com/AceCentre/FaceCommander - dasher. This needs a lot of c++ https://github.com/dasher-project/dasher-MIT/ its the fastest text entry system for head mouse , eyegaze users etc but woefully old

See also openassistive.org makers making change, openaac for other communities

Its BLE needs more work - note that project uses the nrf52840. Its pretty feature complete

..Or iOS, Windows.. or Mac..

Oooh yes. Yes. Let’s chat. We have some ux stuff done by a googler focusing on the config etc. people are desperate to do the zooming bit - I’ve try to hold back the tide until we can do the basics but I’d love some help tying up the designs on the overall app. wwade @ acecentre.org.uk

Ha! That’s me! If anyone wants to help dasher dev please get in touch. We need $$$ or some good c coders willing to spend a couple of months working on it all. It’s good. It’s the fastest text interface for eyegaze, head mouse or “continuous gesture”. We need it desperately for those who can’t speak

I'm with you. I started reading and was like "wait.. this is not the Cheshire I know of - where has this been hiding".. Then on the map: Lebanon and Brooklyn.

Also Wallingford. I bet thats nothing like the Wallingford I know of (Oxfordshire Town)

nice. feedbin also do this https://feedbin.com

a translator: http://www.kafejo.com/lingvoj/auxlangs/dws/tral/index.php

I hire regularly on upwork. There are a TON of developers lying about who they are or where they are in the world (I swear, despite the Russian embargo there are a lot of devs from Russia who have found loopholes via Turkey etc). The biggest problem I have is people trying to get a gig then asking to come off upwork. Don’t do this. You’ll get banned. Happened twice this weekend

But do preserve. We would hire at twice the rate when we can for a better more reliable set of reviews.

and the guy on the right.. umm.. what's with his face? Or is he an Alien maybe? Image credit goes to https://linkmedya.com - it doesn't say it is AI-generating content but yep, it certainly looks like it

Elevenlabs, coqui, piper, Microsoft, Google, Apple. Seriously. They all can these days. Don’t forget acapela or nuance.

I think we could tweak this so you can use it as a morse keyboard/mouse for desktop. I find it interesting the single, double tap thing. Hadnt considered that. https://github.com/acecentre/MorseWriter (nb: if you ever want to develop on this look at https://github.com/AceCentre/MorseWriter/tree/pynput https://github.com/AceCentre/MorseWriter/issues/31)

Their library was actually made for dasher.. http://www.inference.org.uk/dasher/ - there was a web version being made (https://github.com/dasher-project/dasher-web We hit a bottleneck with the graphics driving. Note in dasher pretty much the entire tree is in dynamic view). Now this may help to understand the use case. Dasher is for people with disabilities who cant speak. It needs to be a personalised LM that trains on the fly and and keeps track of new words/sentences. But in truth too, utterances are usually small.

Don't get too knocked back by comments. A) If it works - it works. B) Your learning is as valuable as the outcome.

Oh have a look at https://imagineville.org/software/ for some other things that may be of interest..

nice. Im trying it out. I'm not that bothered personally about integrations (a DOI is really easy to copy and paste from one tool to another). But I'll get in there first.. I switched to Paperpile a while ago (after the death of Citeulike RIP) - and it's pretty darn perfect. I'd vote a +1 to have that as a citation source..

PS is this a electron app? It has that feel..

found the JS one..

https://github.com/google-research/google-research/tree/mast...

I was thinking of “Persistent uniform resource locator” but then realised I think you are referring to the python library https://github.com/codeinthehole/purl (I get your argument although I quite like this new project. It’s closer to “Perl” in my mind. But then I am nearly 50..)

Monkey-Mind. Huh. That’s interesting. Is that kinda ADHD?

Have a read about PPM - Prediction by Partial Matching

http://cotty.16x16.com/compress/peppm.htm

https://compressions.sourceforge.net/PPM.html

https://en.m.wikipedia.org/wiki/Prediction_by_partial_matchi...

In swift https://github.com/kdv123/PPMLM

In C https://github.com/money6g/ppm

Someone’s done a JS version but I can’t find it right now

Interesting. I recently built a non stop drag system for someone who uses it to communicate. The system uses either a dwell or angular detection system to figure out the letters. I wonder though if some path analysis would be better. I wonder if you can reverse this - so from an image of the path (and with time stamp along the path) you can go path -> text.

https://github.com/AceCentre/DragToSpeak see video of it in action https://willwa.de/2024/03/26/iterative-rapid-development.htm...

Thats neat. Thanks. I tweaked my little abbreviation maker tool for that. https://abbreviation-finder.streamlit.app

If you watch the film youll realize it was somewhat based on ideas from pictorial languages - mainly Chinese

NB Michael Everson has been plugging away at the iso standard for a while https://www.unicode.org/wg2/docs/n5149-blissymbols-keyboards...

Simplistic was the aim. Have a watch of the film in the comments above. It was designed as a universal language.

It took off then stagnated a bit. Used a lot in Canada and Iceland, Norway, Sweden and various parts of the globe eg Eastern Europe. There is various projects further developing it still

That book you referenced is cool. But it’s not depicting a true language. Language has rules which allows you to generate new novel words and sentences. This is what bliss does and why it’s called a language - rather than all other symbol systems

Essential listening - https://radiolab.org/podcast/bliss

A nice read https://letterformarchive.org/news/blissymbolics/

And the film https://archive.org/details/mrsymbolman