Back-end built with Python, Pydub, gTTS, SpeechRecognition FastAPI and Pydantic.
If you don't already installed poetry
in your machine, run:
pip install poetry
Initialize and install dependencies with poetry
:
poetry shell
poetry install
To communicate with OpenAI
API is necessary to set up your API_KEY
, so copy our .env
example file:
cp .env.dev .env
Inside the .env
change the variables with your keys.
NOTE: If you don't already have the OPENAI keys, you can create here https://platform.openai.com/account/api-keys
To run the ASGI server:
python -m talk_gpt.server
# or: talk_gpt --cmd api
To run by specifying your environment variables file:
python -m talk_gpt.server --env-file <ENV_FILE>
# or: talk_gpt --cmd api --env-file <ENV_FILE>
Access: http://localhost:8000/docs
To start a chat via CLI:
talk_gpt --cmd cli
To start via CLI listening responses in audio:
talk_gpt --cmd cli.audio
Give a name to your chat and reuse that easy:
talk_gpt --cmd cli --chat-id <CHAT_NAME>
# Then you can always run this command and get your historic
To start with your keys without any configuration:
talk_gpt --openai-key <YOUR_KEY> --openai-org <YOUR_ORG>
For more custom settings:
talk_gpt --help