willwade

17 ms·

1.
by willwade 13d ago
I worked for .name briefly right at the beginning of their entry into the world. Interesting but very odd part of my career.. Ill give it that.. I personally found the product at first a great idea but somewhat crippled by execution..
2.
by willwade 2mo ago
Sony presented a simpler solution to this at CHI https://dl.acm.org/doi/10.1145/3772318.3791397 - check video in that. Also it seems every Cornell student is working on this. Just search on YouTube Cornell silent
3.
by willwade 4mo ago
Interesting. It sounds like something I was doing : https://github.com/Smartbox-Assistive-Technology/app-automat... my use case though is people with disabilities. Not robots.
4.
by willwade 4mo ago
It's been done for a while - follow the links to who they reference. ie https://www.tolt.tech but it's their integration they've done into the OS is interesting.
5.
by willwade 4mo ago
I can promise you it isn’t. This totally predates llms. Been using them for a long time
6.
by willwade 4mo ago
Use those a lot to (always try and focus off the shelf first). But for cases where we need to position differently we often need something custom
7.
by willwade 4mo ago
I’ve provided a lot of different of the balls. Last one was for someone in bed and needed to control the whole pc with his chin. We adapted the base to fit on a mount. Other uses for people with ALS with reduced hand function - altered the
8.
by willwade 4mo ago
I’ll give you a use case. People with disabilities who can’t use the keyboard. Ploopy stuff has been absolutely amazing.
9.
by willwade 4mo ago
I find that the new "drug" is constantly hunting down new cheaper models.. z.ai/glm, mistral, deepseek.. if you need to get your fix - find the cheaper path..
10.
by willwade 4mo ago
take a peek at https://github.com/willwade/app-automate?tab=readme-ov-file#... - its early and needs some work -but this is the idea behind this.. (my use case is not agents but actual real disabled people..who need to
11.
by willwade 4mo ago
Interesting! I started something - nowhere near as complete as that and quite different but again using accessibility UI elements. The BIG problem I've found is SOOOO much stuff does really poorly having these elements exposed. Here wa
12.
by willwade 6mo ago
It gave me strong vibes too. It’s the writing style. I’ve seen OpenAI write just like this. Doesn’t mean it’s bad. There’s a few other markers. Note “silly” in quotes and over use if that word. Once would be enough. But also this is very ve
13.
by willwade 8mo ago
how did you redact the sensitive data? did you use a library for that?
14.
by willwade 8mo ago
this is so cool. I think so much of the logging Saas products are bloated. This looks much simpler. I'd like a python interface.. I've used papertrail and datadog for some of this in the past but dropped it due to cost (and bloat)
15.
by willwade 8mo ago
.. and fentanyl for sure right? surely its the largest producer of fentanyl.. or is it cocaine? hold on whats that drug ICE? ah thats it..
16.
by willwade 9mo ago
sorry.. global gitignore.. what have i been doing..
17.
by willwade 9mo ago
that to me looks like a error in whatever logic is behind the positional error code. You'd think they would have transformer models based on different layouts but maybe some weighting issues going on.. ie I would have thought its a mod
18.
by willwade 9mo ago
I wonder if this would have been useful https://github.com/microsoft/presidio - its heavy but looks really good. There is a lite version..
19.
by willwade 9mo ago
can i have this between my machine and git please.. Like its twice now I've commmited .env* and totally passed me by (usually because its to a private repo..) then later on we/someone clears down the files.. and forgets to rewrite
20.
by willwade 10mo ago
Meta cheated with the mms models. That is they didn’t use a phonemeizsr step. This means they just won’t work or sound very strange. ASR data is usually not quite right for tts. But anyhow - not really answering your question but many of th
21.
Word, Character prediction in any language in the browser (ppmpredictor)(willwade.github.io)
1 points by willwade 10mo ago|1 comments
22.
by willwade 10mo ago
We built this for our use case (we create solutions to help people speak who have a disability). This is a prediction model you can run in node or the browser. Next word, next character, word completion.. PPM is old - but still rocks
23.
by willwade 11mo ago
I’m interested to see how this compares with other heroku clones. The compassion stuff is interesting. I’m using apps on digitalocean. Can we get a comparaison of using app with droplet+blossom? I’m with the other person too. Drop the emoji
24.
by willwade 11mo ago
Yeah I totally want this. How much data are we talking about on average?
25.
by willwade 1y ago
See also https://github.com/mikeborozdin/vibe-composer-midi-mcp?tab=r...
26.
Vibe composing with a MIDI MCP [video](youtube.com)
2 points by willwade 1y ago|1 comments
27.
by willwade 1y ago
I’ve developed a couple of fine tuned t5 grammar correction models. Is the model or training open?
28.
by willwade 1y ago
oops. my bad. I take it back :)
29.
by willwade 1y ago
Their opentext API is actually largely marketing - infact so much it worked - Im going to make some Monster cakes https://opentextapi.monsterenergy.com/opentext/images/ecde50... - https://opentextapi.m
30.
by willwade 1y ago
Your datasets. Are they public? For more under represented languages we DONT need closed voice models - what the world really needs is open voice data repositories (eg TTS ready voice banks AND phonemization db in projects like Mozila Commo

More ›