How to use free options?

[Needthisplz]

what are the .env templates for the free options? Also is there a free Openai alternative?

[JarikDem-Bot]

Hello! Thanks for reaching out and your interest in the project.

As for now, this program supports only OpenAI to generate responses. However, in the future I will add support of alternative services. Though I'm not sure how soon it'll be, as I have a few other major feature in mind, such as GUI and video recognition.

If you're looking for free alternatives, you can choose free speech recognition and text-to-speech services in main.py file in waifu.initialize method. Simply set the following values:

• user_input_service='google' - to select free Google speech recognition

• tts_service='google' - to select free Google text-to-speech service

As for .env file, you'll need to leave only OpenAI Key:

      OPENAI_API_KEY='YOUR_OPEN_AI_KEY'

Feel free to suggest any alternative services you'd like to see supported in the future! If you have any more questions, I'm here to help!

[Needthisplz]

I have another question: where did you learn to create something like this? I’ve tried finding tutorials for specifically this kind of thing but I just can’t learn it. I would love to build my own just to prove I can to some people that mocked me. I’m perfectly willing to put in the work, I’ve no idea where to start though.

[JarikDem-Bot]

Well, when I was in elementary school I started to go on local programming courses, where I got a solid foundation in C/C++. This helped me a lot when I started to learn python. To learn the basics of python, I used an online course on Codecademy, but any other python course for beginners should do. After that, I just continued to learnt it by myself, from YouTube video, blogs, forums, articles e.t.c. and by making some projects.

Also, it really helps if you're working on something specific because you have particular topics you need to learn. So it's awesome that you're willing to build a project like this and ready to put in the work!

And here are some resources that may be useful in creating your own AI assistant:

• Simple python AI chatbot using OpenAI. It's beginner-friendly, has easy to understand descriptions of code and what it does, and good structure. In fact, I used this exact article to make my first program with OpenAI!
• OpenAI Python API – Complete Guide. This one describes how to work with the majority of OpenAI features in python, so it can be a bit overwhelming for a start. So focus on the things you need the most at the moment - introduction and installation, text and chatbot.
• Also, there are a bunch of videos in the YouTube like python ai vtuber, python ai waifu, python openai chatbot and so on. Usually they don't go in depth to describe code, but they describe the general structure of the algorithm. For example: get response from user from console -> generate response using OpenAI -> print the response -> repeat. It will help you to break your program into smaller sections and build them one by one. Here are a few videos about anime AI vtubers that may be helpful:
  • video "How I Programmed My Own AI Girlfriend"
  • video "How I Programmed My AI Vtuber"
  • video "Open Source Twitch AI Cohost - Vivy Enhancements"
  • video "How I'm Adding Karaoke to My AI Vtuber - Vivy"
  • video "How to Make Your Own AI Waifu Virtual VTUBER or Assistant"
  • video "How to make an AI VTuber Using GPT 3 and Google Cloud TTS"

Remember to break project into smaller task and figure it out one by one. It's better to go slowly but surely. For example, start from simple python AI chatbot article. Then add personality to your chatbot. After that, you can figure out text-to-speech. Then add speech recognition and so on.

I hope I could help you with this at least a little! Best of luck with your learning journey! And if you'll have questions or need some guidance, feel free to reach out!

[Needthisplz]

Hey again, so I’ve been searching for alternatives to suggest. I just read about Microsoft’s Lobe. It’s a machine learning platform for building models. Would that work? Please keep in mind that I am just starting off and have to ask in order to learn. Some have gotten upset when I ask “stupid” questions.

[JarikDem-Bot]

Don't you worry, everyone starts somewhere, so it's totally fine to ask questions! But, sadly, it's also true that there are a lot of people who get annoyed when sees a simple question. Don't let anyone discourage you and keep moving forward!

1. So, I've looked a bit into Microsoft Lobe. It is indeed free ML platform. But it can be used only for image classifier. As you can see on the example below, it can be used to distinguish different types of plant by photo.

   

   Considering this, Microsoft Lobe can't be used to create an AI chatbot. But, you can implement it later as one of the features, so that your chatbot can also support image classification.

2. As for free chatbot alternatives, you can try out CharacterAI. This platform has a lot of different AI characters that you can use. Originally, it doesn't have API, that will allow you to create python program with it. But, fortunately, there is a simple way thanks to an unofficial library created by one developer. Here's a few useful links:

   • Video example - not the best video to use as tutorial, but you'll see an example of how it works in action
   • Documentation - describes how to use this module. Also, it has a few examples of possible usage, such as chatting
   • Repository of this module - here you can look into the source code of this library, find link to discord server of this library, star repository if it was useful and so on

   When I'll return to this project, I plan to add support of CharacterAI thanks to the variety of different pre-created characters.

3. Oh, also there's one more option. If you have good hardware, you can run LLM locally (Large Language Model - model, that generates text, ChatGPT is one of the examples of LLM). There are a lot of free options and plenty of tutorials. But, unfortunately, I've never worked with them due to my laptop's limitations, so I won't be of much help here.

   • Tutorial on running GPT LLMs locally

[DOCYEDOC]

Well, when I was in elementary school I started to go on local programming courses, where I got a solid foundation in C/C++. This helped me a lot when I started to learn python. To learn the basics of python, I used an online course on Codecademy, but any other python course for beginners should do. After that, I just continued to learnt it by myself, from YouTube video, blogs, forums, articles e.t.c. and by making some projects.

Also, it really helps if you're working on something specific because you have particular topics you need to learn. So it's awesome that you're willing to build a project like this and ready to put in the work!

And here are some resources that may be useful in creating your own AI assistant:

• Simple python AI chatbot using OpenAI. It's beginner-friendly, has easy to understand descriptions of code and what it does, and good structure. In fact, I used this exact article to make my first program with OpenAI!
• OpenAI Python API – Complete Guide. This one describes how to work with the majority of OpenAI features in python, so it can be a bit overwhelming for a start. So focus on the things you need the most at the moment - introduction and installation, text and chatbot.

• Also, there are a bunch of videos in the YouTube like python ai vtuber, python ai waifu, python openai chatbot and so on. Usually they don't go in depth to describe code, but they describe the general structure of the algorithm. For example: get response from user from console -> generate response using OpenAI -> print the response -> repeat. It will help you to break your program into smaller sections and build them one by one. Here are a few videos about anime AI vtubers that may be helpful:

  • video "How I Programmed My Own AI Girlfriend"
  • video "How I Programmed My AI Vtuber"
  • video "Open Source Twitch AI Cohost - Vivy Enhancements"
  • video "How I'm Adding Karaoke to My AI Vtuber - Vivy"
  • video "How to Make Your Own AI Waifu Virtual VTUBER or Assistant"
  • video "How to make an AI VTuber Using GPT 3 and Google Cloud TTS"

Remember to break project into smaller task and figure it out one by one. It's better to go slowly but surely. For example, start from simple python AI chatbot article. Then add personality to your chatbot. After that, you can figure out text-to-speech. Then add speech recognition and so on.

I hope I could help you with this at least a little! Best of luck with your learning journey! And if you'll have questions or need some guidance, feel free to reach out!

Can I change tts engine I wanna change to openai tts

[JarikDem-Bot]

Unfortunately, currently Waifu supports only Google TTS service and Elevenlabs. This project is on hiatus, so I am not sure when I'll be able to add OpenAI TTS service