Comments (3)
Hey no probs!
Yes this kind of thing was on my hit list, but I've wanted to sit back a little and just make sure things are stable for a while as I was firing out code every few days.
However, I've done a quick code update for the TTS generator, so you are welcome to pull an update. What I've set it to do is pull these variables from the actual browser window, so as you have to load the page from the AllTalk server, be it IP address, dns address or whatever, it will always point back to the location of the AllTalk server.
var protocol = window.location.protocol;
var host = window.location.hostname;
var port = window.location.port;
var baseUrl = protocol + "//" + host + (port ? ':' + port : '');
So we don't have to account for FastAPI or anything like that. All the functions will just point to the IP/name the page is loaded from and you have to run the page from AllTalk, webserver, so this should render a universal solution.
As for changing the IP, you should do that through the http://127.0.0.1:7851/ settings page as that will change the confignew.json file for you, but even then only if you need to explicitly serve it on an specific IP I guess, you are correct though, you cant bind it to 0.0.0.0 without using a DNS record or name based routing in some way, back to a specific IP. I'm sure I can pop something in the documentation somewhere. (BTW I'm just adding this detail for anyone else who reads this issue).
The only thing of AllTalk's that's running gradio is the finetuning. Whist it does integrate into the Gradio of Text-gen-webui thats a locally (python) interaction that's occurring there, nothing over any IP's/network. Is it a concern with the Finetuning you are thinking about?
Thanks
from alltalk_tts.
Wow thanks for the quick fix. I had not considered such a simple JS fix.
Regarding configuration, your suggested method won't work without some hoop-jumping if you are running on a remote machine. How can you access the settings page when its only available over local host? (note: I know you can do some port forwarding jazz to do this, but again from reading through the docs it was unclear what was needed).
I think it's a valid usecase to be using a remote machine for AI related work, I have a machine stacked with quadros, but I only turn it on when I want to use something on those. At any rate, it is running headless in a basement so without knowing you have to do port forwarding or other such tactics to do the initial config it was a bit of a lift to get up and running
RE: gradio. I was only referencing that as I saw some gradio code. I have no specific concerns with the fine-tuning bits. Hacking on my own Gradio stuff I set the listen port to 0.0.0.0
which was what I was assuming should be happening here. Your explanation cleared that up however.
I'm happy to pitch in somewhere if you have a list of things are are looking for a contributions. This project with the api has the potential to save me some $$ as I won't have to use OpenAPI for what I was looking to do. Happy to see you succeed
from alltalk_tts.
Yeah, I get what you mean with headless systems. I've popped a note in here and then added basic instructions here
I think that will cover 99% of people running a headless system, as that's not the sort of thing your average user will have setup, so I'm assuming they will have a basic level of technical understanding (if they can get a headless system running).
I'm happy to pitch in somewhere if you have a list of things are are looking for a contributions. This project with the api has the potential to save me some $$ as I won't have to use OpenAPI for what I was looking to do. Happy to see you succeed
Well that's great on all counts! :) As I mentioned, I've taken a slight step back for a week or so, because I wanted to reach a point of stability, have people move to up to a code version where I know everything is working without issue, tweak any small changes and have a think about what next. I've actually yet to announce on Reddit/Ooobabooga that I've made a load of changes and any day soon SillyTavern will update and then the AllTalk extension for SillyTavern may end up being used by a new audience, so I'm slightly bracing for that in case I get an influx of questions.
However I'm always up for assistance/help/ideas etc. I do keep a loose list of things people ask me for here #74 and obviously I have other ideas rolling around my head. I guess the only thing is I don't want to do is end up with multiple people working on things that overlap/conflict. But if you're interested in working on something that captures your interest and/or you have some ideas for bits you think would improve it, happy to chat them over discuss a general direction.
For now Im going to mark this ticket as closed, but feel free to reply back here, or open a new conversation on the Discussions area.
Thanks
from alltalk_tts.
Related Issues (20)
- Gradio Off HOT 2
- Issue with Czech sound generation HOT 1
- On error not closing logs HOT 1
- New Tokenizer Loading HOT 24
- Request: Silent installer HOT 5
- Continuing where you left off. No Training effect. HOT 1
- How to further train a successfully trained model? HOT 3
- Can you release pre-compiled latest DeepSpeed for Python 3.10? I see only for Python 3.11 HOT 4
- Stream request not working on ios devices or on Safari HOT 1
- Support for MARS models HOT 3
- long load time model HOT 3
- is it possible to add support for Styletts2 ,Gpt-Sovits and EmotiVoice, HOT 1
- Handling Concurrent Streaming Requests in My Project HOT 1
- [GEN] Error during audio generation: 500: Exception while running subprocess: [Errno 2] No such file or directory: 'piper' HOT 1
- problem with Polish language HOT 7
- Standard TTS generation example from the documentation does not work HOT 4
- Need help :( HOT 1
- Support for metavoice-1B HOT 1
- Are you going to add latest Parler TTS large v1 and mini V1 models? HOT 3
- Loading Parlor at 16-bit HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from alltalk_tts.