adi-panda / kuebiko Goto Github PK

Hi, please I was wondering if I can lock the process so my avatar would only speak when asked, I can't understand or find the solution for this, maybe is only adding an if, but I'm not good at it.

hello, I want chatgpt to play various roles and include simple action instructions in the generated responses. Then the VTuber takes action according to the instructions. like:

me: can you turn around?
chatgpt: yes, i can (turn around body).
vtuber voice: yes, ican.
vtuber action : turn around

Hi was following the video closely currently stuck at 10:37

After executing "pip install -r requirements.txt", a bunch of the requirements started to be downloaded. I've included the error below.
I appreicate any help I can get on this.

"
Using cached numpy-1.24.1.tar.gz (10.9 MB)
Installing build dependencies ... done
Getting requirements to build wheel ... error
error: subprocess-exited-with-error

× Getting requirements to build wheel did not run successfully.
│ exit code: 1
╰─> [33 lines of output]
Traceback (most recent call last):
File "C:\Users\redacted\miniconda3\Lib\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 353, in
main()
File "C:\Users\redacted\miniconda3\Lib\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 335, in main
json_out['return_val'] = hook(**hook_input['kwargs'])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\redacted\miniconda3\Lib\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 112, in get_requires_for_build_wheel
backend = _build_backend()
^^^^^^^^^^^^^^^^
File "C:\Users\redacted\miniconda3\Lib\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 77, in build_backend
obj = import_module(mod_path)
^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\redacted\miniconda3\Lib\importlib_init.py", line 90, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "", line 1387, in _gcd_import
File "", line 1360, in _find_and_load
File "", line 1310, in _find_and_load_unlocked
File "", line 488, in _call_with_frames_removed
File "", line 1387, in _gcd_import
File "", line 1360, in _find_and_load
File "", line 1331, in _find_and_load_unlocked
File "", line 935, in load_unlocked
File "", line 995, in exec_module
File "", line 488, in call_with_frames_removed
File "C:\Users\redacted\AppData\Local\Temp\pip-build-env-w6oo9_xy\overlay\Lib\site-packages\setuptools_init.py", line 16, in
import setuptools.version
File "C:\Users\redacted\AppData\Local\Temp\pip-build-env-w6oo9_xy\overlay\Lib\site-packages\setuptools\version.py", line 1, in
import pkg_resources
File "C:\Users\redacted\AppData\Local\Temp\pip-build-env-w6oo9_xy\overlay\Lib\site-packages\pkg_resources_init.py", line 2172, in
register_finder(pkgutil.ImpImporter, find_on_path)
^^^^^^^^^^^^^^^^^^^
AttributeError: module 'pkgutil' has no attribute 'ImpImporter'. Did you mean: 'zipimporter'?
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error

× Getting requirements to build wheel did not run successfully.
│ exit code: 1
╰─> See above for output.

note: This error originates from a subprocess, and is likely not a problem with pip.
"

It is an awesome project!

I am trying to implement a powerful AI-VTuber and I find this amazing repo.

But if we use the GPT-3 API, we can not fine-tune the model with the chat history.

Though, we could add the history to the prompt text.

It is still a limitation for handling a long chat history.
(OpenAI will limit the length of context)

In my opinion, we should keep a local trainable model for archiving a more real AI VTuber?

apparently its missing a whole lot of data for languages that is not english, is there a way to make her words catalogue for other languages more vast?

hello, i got this error

mmdevice audio output error: cannot initialize COM (error 0x80010106)

any solutions ?

"Hi, I encountered an error when running the command 'python main.py' for the Twitch chatbot. Specifically, I got a 'FileNotFoundError' saying that the 'libvlc.dll' module could not be found in the directory specified. I checked my setup and it seems like the module is indeed missing. Can someone please advise on how to resolve this issue? Thank you!"

The method I have tried but not useful (same error):

1. reinstalling the vlc library by running the following command in your command prompt:
   pip uninstall python-vlc
pip install python-vlc
2. manually installing the vlc library by downloading it from the website https://www.dll-files.com/libvlc.dll.html for Version 3.0.8.0 into "Kuebiko-main" folder

I tried running main.py after debugging some code to update with openai's latest version. It gives this error:
Info: Sending Packet with ID 22320
FATAL: Exception in main loop: [WinError 1225] The remote computer refused the network connection

it replies on the first prompt then ignores after that.

Hello, I'm thinking of tuning the AI ​​to do what I want but I don't know how or where to put the final model in the code. Does anyone know how to help me?

Users want access to as much AI as they can get, they dont want to manage 50 accounts, they want the fastest AI they want the cheapest AI, and you can provide all of that for them with this update.

in addition to or in place of integrating with any aggregators - Please integrate APIpie so devs can access them all from one place/subscription and plus it also provides:

-The most affordable, reliable and fastest AI available
-One API to access ~500 Models and growing
-Language, embedding, voice, image, vision and more
-Global AI load balancing, route queries based on price or latency
-Redundancy for major models providing the greatest up time possible
-Global reporting of AI availability, pricing and performance

Its the same API format as openai, just change the domain name and your API key and enjoy a plethora of models without changing any of your code other than how you handle the models list.

This is a win win for everyone, any new AI's from any providers will be automatically integrated into your stack with this one integration. Not to mention all the other advantages.

How I can add my custom fine tuned model? this code triggers me error, I don't know why

chat.py
import openai

def open_file(filepath):
with open(filepath, 'r', encoding='utf-8') as infile:
return infile.read()

openai.api_key = "mykey"
openai.api_base = 'https://api.openai.com/v1/chat'

def gpt3_completion(messages, engine='davinci:ft-personal-2023-04-07-21-57-02', temp=0.9, tokens=400, freq_pen=2.0, pres_pen=2.0, stop=['DOGGIEBRO:', 'CHATTER:']):
response = openai.Completion.create(
model=engine,
messages=messages,
temperature=temp,
max_tokens=tokens,
frequency_penalty=freq_pen,
presence_penalty=pres_pen,
stop=stop)
text = response['choices'][0]['message']['content'].strip()
return text

I would like to avoid having to read the chat icons all the time. How can I implement this?
there is a way to filter chat icons?

Does anyone know how to implement gpt-3.5 to the code? seems like a better and cheaper option for this project than previous openai models. Thank you.

Hi, I would like to add a word filter. Is it possible?
I have read the OpenAI documentation about moderation, but I don't understand how to implement it.
Can someone help me? Thank you in advance.

[0000025d41d7fae0] mmdevice audio output error: cannot initialize COM (error 0x80010106)

Hi, I have had a great time with this project! I was wondering if there was a way to replace the Open AI response fetch with a Stanford Llama model that I would locally run on a networked computer. That would be class because it would allow me to create my own filter allowing the Vtuber to be a bit more risque than what Open AI allows. Thank you.

I know I am cramming the box and I'm sorry about that. I do want to get out the things I would like to see in future versions of this project. I have a few feature ideas that would probably require a true/false toggle. I came across the desire for these features as I am using the AI VTuber as a co-streamer.

• I would like it to be able to be listening to closed captions to activate conversations with me. I have closed captions on my mic and I can also get it to work for discord too through OBS to have twitch display, I just have to tell the captions what channel to listen to. Ideally, it would listen when it's name was mentioned and then listen for 30 seconds after the last response to continue a conversation and then turn itself off if there is 30 seconds of silence. Maybe 30 seconds is too long but I am kind of spitballing here with that.

• I would also like it to start it's own small talk conversations if possible when there has been no action on closed caption for like a minute. Maybe tell a story about it's day or something that happened to them that was funny. I mean I know part of that belongs in the prompt but it would need to be programmed in to activate those kind of conversations.

• I also had a crazy idea of the prompt temperature changing to match the "mood" of the conversation. If people are being mean or she doesn't like the conversation her responses are more rigid and if she is in a "good mood" or people are just being nice, she can give more loose and unpredictable responses. I noticed that she has positive and negative responses, perhaps after the response if it contains negative words, it does a -.1 and if it has positive words in the response it's +.1 otherwise it is 0 on the change. I'm sure of all three, this one would be the hardest to execute so, you can tell me to get bent on this one lol. It would be cool if it could though.

If you ever want to audit what I'm doing, drop by my stream sometime.

Chracter Mouth open and follow according to the response but sound is not coming

If I wanted the bot to thank new followers as they come in, and not only respond to chats, how could I do that?

After a while, because of all the conversations going on. The AI forgets about the initial information set by prompt_chat.txt.

Is there a good way around this?

I tried adding a very makeshift way of pushing the prompt in again after every conversation, but that just made me get token errors, because it's adding too much to it. I'm absolutely doing it the wrong way, because I'm not too good at this stuff.

 async def event_message(self, message):
        ...
        ...
        ...
        Bot.conversation.append(
            {'role': 'system', 'content': open_file('prompt_chat.txt')})
        print('------------------------------------------------------')
        os.remove(audio_file)

        # Since we have commands and are overriding the default `event_message`
        # We must let the bot know we want to handle and invoke our commands...
        await self.handle_commands(message)

raise self.handle_error_response(openai.error.InvalidRequestError: This model's maximum context length is 4097 tokens. However, you requested 4184 tokens (3784 in the messages, 400 in the completion). Please reduce the length of the messages or completion.

I'm basically trying to find a way to inject the prompt chat info now and then, without it ruining everything. Piling it up or what have you. Because clearly the terrible way I'm going about it, is not working.

Thanks for this project by the way, truly appreciate it. Having so much fun with it.

hello, i really like your work!
but im struggling to figure out how to make everything work..
is there a way you can make a video tutorial on how to setup everything and make this work?

How would I get it to talk randomly to fill dead silence and talk about how their day was, tell a story, etc. and not just only rely on Twitch chat?

I would appreciate if someone could help me with this!

Hey! saw there was some updates on the master branch so tried to grab them - Couldn't find the google application credentials area - was also getting this error -

File "C:\Users\jack\Desktop\AI Streamer 4\Kuebiko-master\main.py", line 3, in
from chat import *
File "C:\Users\jack\Desktop\AI Streamer 4\Kuebiko-master\chat.py", line 10, in
with open(os.path.expanduser('~') + '/.kuebikoInfo.json') as f:
FileNotFoundError: [Errno 2] No such file or directory: 'C:\Users\jack/.kuebikoInfo.json'

I live in Russia, and I would like to do this, but I can’t, Google doesn’t provide this option, and there are a lot of problems with this application, can you help me please?

Hey,

I don't suppose you could rework this to instead of being for twitch, it is integrated with Discord.

As in....

It is linked to a bot that we can create in the Discord Dev Portal, it joins a channel when the Host joins and it uses Speech recognition to listen to what that person says and then responds text to speech.

This would be extremely helpful if you have the time to look in to this.

Thank you.

can someone implement azure tts or something that's better, thanks in advance

Is it possible to change to local free model such as Vicuna/LLama/Alpaca/GPT4All?

Hey, there is another AI Vtuber project out there and it's pretty decent asides from it having Japanese translations that seem to bulk it up for an english speaker. I was hoping maybe you could get some inspiration from them to apply to your project:

https://github.com/ardha27/AI-Waifu-Vtuber

Maybe a little freer and saying inappropriate things from time to time, it becomes boring that she is always correct and respectful

can you please provide alternative to google cloud api.