fintech-chatbot

To run the Python program for speech-to-text, first install the following:

sudo apt install ffmpeg
sudo apt install libespeak1
sudo apt install portaudio19-dev
sudo apt install python3-tk
sudo apt install python3-pil python3-pil.imagetk

Next, install the packages in the requirements.txt file, as follows:

pip install -r requirements.txt

Next, provide an OpenAI API Key, as follows:

export OPENAI_API_KEY="<OpenAI API Key>"

In the Python program file, fill-in the details. These can be obtained from the SingleStore portal.

s2_user = "<username>"
s2_password = "<password>"
s2_host = "<host>"
s2_port = <port>
s2_db = "<database>"
db = SQLDatabase.from_uri(
        f"singlestoredb://{s2_user}:{s2_password}@{s2_host}:{s2_port}/{s2_db}"
        "?ssl_ca=/path/to/singlestore_bundle.pem"
)

The singlestore_bundle.pem file is required if using the Free Shared Tier, otherwise this line can be removed:

"?ssl_ca=/path/to/singlestore_bundle.pem"

Finally, run the Python program, as follows:

python3 record-transcribe-visualise-speak-full.py