https://github.com/atomlayer/llama_cute_voice_assistant
Llama cute voice assistant
https://github.com/atomlayer/llama_cute_voice_assistant
assistant llama llama2 oobabooga rvc silero whisper
Last synced: 4 months ago
JSON representation
Llama cute voice assistant
- Host: GitHub
- URL: https://github.com/atomlayer/llama_cute_voice_assistant
- Owner: atomlayer
- License: agpl-3.0
- Created: 2023-08-14T18:39:30.000Z (almost 3 years ago)
- Default Branch: master
- Last Pushed: 2023-09-10T21:38:49.000Z (almost 3 years ago)
- Last Synced: 2026-03-06T05:43:00.924Z (4 months ago)
- Topics: assistant, llama, llama2, oobabooga, rvc, silero, whisper
- Language: Python
- Homepage:
- Size: 573 KB
- Stars: 27
- Watchers: 3
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# llama_cute_voice_assistant
Motivation: Many voice assistants with artificial intelligence have an unpleasant voice. While it may be a matter of personal preference. I aim to create a more human assistant that allows you to plug in cute anime character voices and vtuber voices.
Solution diagram:
# How to install
*This guide may not be the most detailed. It will need to be improved.*
## Step 1 - Install llama_cute_voice_assistant
*Installation procedure for Windows*
1) Install git https://git-scm.com/downloads
2) Install cudatoolkit (You should only choose versions of cuda that pytorch supports. See supported versions here https://pytorch.org/get-started/locally/)
3) Install miniconda https://docs.conda.io/en/latest/miniconda.html
4) Open miniconda console
5) Create new conda environment
```
conda create --name llama_cute_voice_assistant python=3.11
```
6) Activate conda environment
```
conda activate llama_cute_voice_assistant
```
7) Clone project
```
git clone https://github.com/atomlayer/llama_cute_voice_assistant.git
```
8) Go to project directory
```
cd llama_cute_voice_assistant
```
9) Install pytorch
- Go to https://pytorch.org/get-started/locally/
- Generate a command to install pytorch for your system (the command will be like this: pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118 )
- Execute the command
10) Install the libraries
```
pip install SpeechRecognition==3.10.0
pip install pyttsx3==2.90
pip install soundfile==0.12.1
pip install simpleaudio==1.0.4
pip install pygame==2.5.1
conda install PyAudio
pip install openai-whisper --no-cache-dir
pip install omegaconf==2.3.0
pip install git+https://github.com/openai/whisper.git
conda install -c conda-forge ffmpeg
```
## Step 2 - Install oobabooga Text generation web UI
1) [https://github.com/oobabooga/text-generation-webui\#one-click-installers](https://github.com/oobabooga/text-generation-webui#one-click-installers)
2) Open the oobabooga Text Generation web UI using the **-api** parameter.
3) On the model tab: download and run your favorite AI model.
4) On the Chat settings \> Character tab: set your character name and description.
## Step 3 - Install Realtime Voice Changer
Detailed instruction -
1) Download https://huggingface.co/wok000/vcclient000/blob/main/MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.11.zip
2) Unpack the archive
3) Run start_http.bat
4) Join the AI Hub Discord: https://discord.gg/aihub
5) Go to the search-models channel
6) Find and download the model you like
7) Click the edit button in the Realtime Voice Changer Client
8) Upload the model to a free cell
9) Adjust the best TUNE parameter for your voice.
## Step 4 – Install the virtual microphone
1) Download and install VB-CABLE Virtual Audio Device
2) Open Realtime Voice Changer
3) Set up an audio input: Cable Output (VB-Audio Virtual Cable)
4) Press start button
## Step 5 - Change the settings in the settings.py file
1) Replace oobabooga_api_name and wake words.
oobabooga_api_name - the name of one of your characters in oobabooga text generation web UI (Paremeters > Character tab)
```
oobabooga_api_name = "Lisa"
wake_words = ["lisa"]
```
# How to use
1) Open conda console in the project folder
2) Run the command:
```
python voice_chat.py
```
3) Say the wake word and the command for your assistant.
# Links
https://github.com/davabase/whisper_real_time