Comments (5)
Let me clarify on your request. You are asking to generate the same text, but in 2x different voices. And then you want to be able to merge/nit portions of the generated TTS back together to generate the final exported WAV file, where different sections would be read by different voices (as you choose)?
from alltalk_tts.
from alltalk_tts.
Now you're up and running with this, back to this question. So each time you run a generation, the files created are unique, so if you ran one generation with one lot of text and the voice of your choosing, you would have files A, B, C....... etc
and you can save that off as a JSON list or Export to a combined wav file.
If you clear the list and changed the text and/or voice, or even just even left the same text and voice in there, but hit generate again, you will get completely new files generated that will sit alongside the existing files, so 1, 2, 3.... etc
. Again you can export that list to JSON or to a combined wav file.
In the output folder you would have files A, B, C...etc 1, 2, 3..... etc
. Until you delete the originally generated wav files, you can just re-load the JSON files to regenerate individual bits or export the list to a combined wav file.
Now theoretically speaking (not that Ive tried this) If you open 2x web pages and set one off with voiceA
and textA
and you set the off with voiceb
and textb
, this would result in a race condition I believe where it will shuffle the 2x generations together, due to the way each web page is waiting for a JSON response. Is it possible to work around this in future, probably yes, it may be as simple as adding a unique ID to each generation request and checking for it on each return OR it may be a very large re-write of the API, generator and other such things. It would be something I would have to think on.
You can however, as I mention at the start, just create the 2x generations individually and still have access to edit, change, regenerate parts as needed and export their individual list to JSON and re-load that (until you delete the original files). Does that at least cover off your basic request a partial solution?
Thanks
from alltalk_tts.
from alltalk_tts.
Ok, so you are basically saying you want to be able to throw multiple texts at it, as raw text or text files of some sort and have them each generate based on settings you set for each document.
This is possible, but due to web-browser limitations, this would more than likely be a very large re-write. Not just the web interface, but potentially the API suite and other things. Or I write a whole separate python script/gradio maybe to handle this. I've added a note on here as something that's been requested, but being such a large thing, it may be a version 2 thing.
from alltalk_tts.
Related Issues (20)
- Cookie Blocker Causing Problems? HOT 1
- Streaming mode not working on Firefox HOT 2
- (Support) Streaming to Unity HOT 2
- There's a problem with the bulk generator. HOT 2
- Feature request: I would like a bulk generator checkbox to also split on new lines. HOT 3
- Impossible to install CUDA Toolkit - Docker HOT 4
- Crash. HOT 5
- AllTalk v1.9c: DeepSpeed Installation Error HOT 5
- Folder path with space in it doesnt handle error well HOT 2
- Headless, invisible mode HOT 2
- Add torch in PATH for docker environment HOT 3
- Can you do an inference demo on Huggingface? HOT 1
- Issue: 'pip' Not Recognized When Installing AllTalk in Text-Generation-WebUI Environment HOT 10
- 🚧 Breaking Changes
- Thank you HOT 1
- Alltalk Unity Integration? HOT 4
- torch index URL is truncated. HOT 1
- Config page provides no way to load a fine-tuned model HOT 5
- Add UI control for the learning rate on the finetuning page. HOT 2
- Feature request: Compatiblity with Home Assistant using Wyoming protocol 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.