As for what to talk to.. or the ultimate purpose lol.. you could do simple device control scenarios ("lights off"), do hands-free sensor readings ("temperature at 95 degrees"), change wifi settings via voice ("switch access points"), etc.
I use it to control an agent-powered diverse device sensor network in my home and my truck. Much of the capability needs Internet access, but having on-device speech to text and text to speech means I can still do some stuff when I didn't bring the Starlink with me. On-device STT opens up a lot of low bandwidth (lora) opportunities too.
Many ESP32S3 dev boards have a microphone and speaker header available conveniently on the board, so I have been using that to respond via Babytalk's text to speech capability.
I'll improve the README to make it a little more clear.
https://github.com/tlack/babytalk
I have been playing with a lot of $20 devices with no keyboards or screens. I wanted to use speech to interact with them, but without it having to run through a cloud service or use a fixed command vocabulary.I worked with Claude to implement high performance 4bit int kernels for the ESP32S3 and P4 and quantize some larger STT and TTS models. Surprisingly, it's quite usable on these tiny...
(I have no affiliation with this company aside from being a happy customer the last few years)
Runpod also has an offering in this space but I think they’re shifting toward more managed-by-them rental services.
If you want someone to take you up on that offer you gotta specify a % differential from cloud list price. :)
Here are some videos related to his work: https://www.youtube.com/playlist?list=PLDU0iEj6f8duXzmgnlGX4...
Co-dfns was most recently discussed on Hacker News 3 months ago: https://news.ycombinator.com/item?id=40928450
On our [1] chatbots we use one long memories text field per chatbot <-> user relationship.
Each bot response cycle suggests a new memory to add as part of its prompt (along with the message etc)
Then we take that new memory and the existing memories text and feed it to a separate “memory archivist” LLM prompt cycle that’s tasked with adding this new memory and resummarizing the whole thing, yielding a replacement for the stored memories, with this new memory added.
Maybe overly simplistic but easy to manage and pretty inexpensive. The archiving part is async and fast. The LLM seems pretty good sussing out what’s important and what isn’t.
But, bad or good, I think all you can really do is keep trying! Don't let a few rejections stymie your long term goals. Your family needs you to keep putting one foot in front of the other and applying to more and more places until you find that perfect role.
Maybe use this downtime to build yourself up: open source some stuff that defines you as a subject matter expert, or blog about some of your experiences, etc.
Wouldn't hurt to share your resume here too - lotta industry people lurking. :)
https://github.com/AUTOMATIC1111/stable-diffusion-webui/blob...
you can try blending the before and after with PIL but it will probably look even more disturbing
strength = 0.5
horrific_image = Image.blend(image_with_normal_skin_and_teeth,
kinda_awful_gfpgan_image, strength)
Worth a shot!
But personally I’ve been working at higher levels of the stack the last few years, where these kinds of decisions seem less important.
And on another level, it seems like coders in general aren’t that interested in vector oriented languages and techniques which makes their study somewhat isolating.
On the Stable Diffusion side: labeling, best practices, file management, quality control
A good prompt engineer understands the scope of keywords trained into SD and can translate the artistic vision for a finished image into specific keyword constructions.
Try creating a specific visual idea you have in your mind into a jpeg using Diffusion. You’ll see it’s not easy.
Can the user do that? Sure! Just memorize all the critical artists of the last 400 years and their important works. And learn all the specific terminology that yields great results.
Can a computer do it? Sure! Just fill out dozens of form fields to describe your specific intent.
I think it's at least worth learning because of the way it makes you think about coding.