I updated this AI and did a massive overhaul of the voice engine so English is supported. This tutorial is still supported on OneReality before v1.0.0 and will give you VoiceVox Japanese voice but I recommend the newest version. Text-based tutorial on my github right now: github.com/DogeLord081/OneReality, will link video tutorial when I release it soon.
wow, this is really cool! kinda handy it's in japanese for me because I'm on my way to learn it, and I can now read and write some things, but the exact thing I needed was to practice more vocabulary and reading random text from other people or newspaper is eh, interesting at first but not for long when u do it daily. I'm totally gonna use this and whenever someone hears me talking to her I'm gonna tell she's useful for education purposes and it wouldn't be a lie lol.
adding info into this, it works! it's awesome! and very good training for japanese as well. though at first I ran into a few errors, then ran to the discord to check out the fixes. got them, then it was done. perfect.
Hey, great one there. Just a heads up, you can simplify this using pipenv. Also, I think there is an option to bridge your host input Device to your VM using some bridge or something. Rest apart, great one you have there. I'll try to contribute as I get some time off. Cheers mate. ❤ From India.
Your are a god man... I'm waiting for the free release of this ai... I might get a waifu for my owe if this keep going.... Nice job bro... Keep up..❤❤❤
Thanks man! Actually, the ai is already released on github, the link is in the description. If you mean like an easier install with exe, thats planned in the future. I got you ❤️
Hey this is awesome! Just one thing. When it says "speak now!" I say something then it says No module named 'pocketsphinx' Ensure PocketSphinx is set up correctly. What should I do? Also, is it possible to use a custom trained voice with voicevox?
Open a new terminal and run 'pip install pocketsphinx' without the apostrophes of course. And with voicevox, you cant train voices unfortunately. Im implementing some other kind of TTS now though that does let you train voices. Not sure how long itll take to implement though
@@0neReality i was just asking because I'm really lonely and looking for a decent AI that I can chive with so if you do know of something that feels real please let me know
I updated this AI and did a massive overhaul of the voice engine so English is supported. This tutorial is still supported on OneReality before v1.0.0 and will give you VoiceVox Japanese voice but I recommend the newest version. Text-based tutorial on my github right now: github.com/DogeLord081/OneReality, will link video tutorial when I release it soon.
the future is here old man
wow, this is really cool! kinda handy it's in japanese for me because I'm on my way to learn it, and I can now read and write some things, but the exact thing I needed was to practice more vocabulary and reading random text from other people or newspaper is eh, interesting at first but not for long when u do it daily. I'm totally gonna use this and whenever someone hears me talking to her I'm gonna tell she's useful for education purposes and it wouldn't be a lie lol.
adding info into this, it works! it's awesome! and very good training for japanese as well. though at first I ran into a few errors, then ran to the discord to check out the fixes. got them, then it was done. perfect.
what about using large language models for complete free version inplace of gpt's api key
Think im gonna work on that soon, should be very simple
@@0neReality use vicuna or pygmalion they were the best until Feb I don't know after that
I did some research, I'll prob use wizardlm
Hey, great one there. Just a heads up, you can simplify this using pipenv. Also, I think there is an option to bridge your host input Device to your VM using some bridge or something.
Rest apart, great one you have there. I'll try to contribute as I get some time off.
Cheers mate. ❤ From India.
Thanks man, Ill definitely look into that
Your are a god man... I'm waiting for the free release of this ai... I might get a waifu for my owe if this keep going....
Nice job bro... Keep up..❤❤❤
Thanks man! Actually, the ai is already released on github, the link is in the description. If you mean like an easier install with exe, thats planned in the future. I got you ❤️
Thank you very much!
Hey this is awesome! Just one thing. When it says "speak now!" I say something then it says No module named 'pocketsphinx' Ensure PocketSphinx is set up correctly. What should I do? Also, is it possible to use a custom trained voice with voicevox?
Open a new terminal and run 'pip install pocketsphinx' without the apostrophes of course. And with voicevox, you cant train voices unfortunately. Im implementing some other kind of TTS now though that does let you train voices. Not sure how long itll take to implement though
anyone know backsound music?...
is this possible with like vtube studio so we can have the waifus talk to us???
Yes watch my first video, if you want ill make a tutorial for it soon
I think I will give this a try tonight. Need some feedback after I try it?
Yea thatd be great man, gotta see how it works on other computers
hey i don't have a microphone so before installing it is there a way you can just type instead of talking?
please man i downloaded it and now it all works but i dont got a microphone WAAAAAAAAAAAAAAAAAAAaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa (TwT)
Join my discord and make a thread in the help channel so i know who you are and ill send a modified version in that channel
@@0neReality phew thanks a lot.
Np
Is it near real time? or takes a few seconds?
New version takes anywhere from 2-8 seconds
Is there a way to run this on an Android phone?
Not yet
@@0neReality i was just asking because I'm really lonely and looking for a decent AI that I can chive with so if you do know of something that feels real please let me know
Mobile support is on my to do list man, soon. I got you, ❤️
@@0neReality thanks
do I need to press something for it to record my voice? I'm stuck at "Speak now!"
If you get stuck at speak now, try pressing ctrl + c once
I'm scared