Chatty the vocal Cheshire Cat API Client in python

This project is a Python client for connecting to a WebSocket API using the cheshire_cat_api module. The client is configured to send and receive messages, including audio files, via WebSocket.

Best use with Local Whisper Cat and Piper Cat for going full vocal and local.

Requirements

Ensure you have the following modules installed:

• datetime
• json
• time
• cheshire_cat_api
• pprint
• base64
• beautifulsoup4
• playsound
• threading

You can install any missing dependencies using pip:

pip install datetime json time pprint base64 beautifulsoup4 playsound threading

Configuration

The client uses a default configuration, but you can modify the parameters as follows:

config = ccat.Config(
    base_url="localhost",
    port=1865,
    user_id="user_boy",
    auth_key="",
    secure_connection=False,
)

Recording and Sending Audio Messages 🎤

The client can record audio messages and send them to the server:

while True:
    message = input("Write 'stop' to exit: \nPress Enter for recording")
    if message == "stop":
        break
    else:
        my_audio = record_audio()
        extra_fields["audio_key"] = read_file_as_base64(my_audio)
        cat_client.send(message, **extra_fields)

Listening the response 🔉

If there is a message with an audio inside the playaudio function will play the response! In the following response format it will look for the audio file

{
    "type" :
        "chat",
    "content" :
        "<audio controls autoplay><source 
        src='/admin/assets/voice/voice_20240617_140731.wav' 
        type='audio/mp3'
        >
        Your browser does not support the audio element.</audio>"
    }

with the function on_message() on a separate thread

 if message["type"] == "chat" and "type='audio/mp3" in message["content"]:
        html_string = message["content"]
        url_audio_to_play = get_audio_file_path(
            html_string
        ) 
        print("The file you are looking for is at :--->" + url_audio_to_play)
        thread = threading.Thread(target=playsound, args=(url_audio_to_play,))
        thread.start()

Running the Client

To run the client, ensure the WebSocket server is up and running, then start the client:

python chatty.py

it will wait for your "enter" and than rec 10 seconds of audio,replying you what you said , the text and audio response

# Connect to the WebSocket API
cat_client.connect_ws()

while not cat_client.is_ws_connected:
    time.sleep(1)

# Run the main loop
while True:
    message = input("Write 'stop' to exit: \nPress Enter for recording")
    if message == "stop":
        break
    else:
        my_audio = record_audio()
        extra_fields["audio_key"] = read_file_as_base64(my_audio)
        cat_client.send(message, **extra_fields)

# Close the connection
cat_client.close()

Main Functions from cheshire_cat_api module

Connection

The WebSocket connection is managed by the CatClient class from cheshire_cat_api with callbacks for specific events:

• on_open: Triggered when the connection is opened.
• on_message: Triggered when a new message arrives.
• on_error: Triggered when a WebSocket error occurs.
• on_close: Triggered when the connection is closed.

Message Handling

The on_message method handles incoming messages, looking for audio files in the message content and playing them if found.

File Reading and Encoding

The client includes functions to read files and encode them in base64:

• read_file(file_path): Reads a file and returns its content.
• read_file_as_base64(file_path): Reads a file and encodes its content to base64.

Extracting Audio File Paths

The get_audio_file_path method extracts the audio file path from an HTML string and returns the complete URL to reach the audio file.

Contributions

If you wish to contribute to this project, feel free to fork the repository and submit a pull request.

License

This project is licensed under the GNU General Public License v3.0..