https://github.com/lexiestleszek/talkr
Local AI assistant to talk about stuff (using Ollama and Whisper)
https://github.com/lexiestleszek/talkr
Last synced: 3 months ago
JSON representation
Local AI assistant to talk about stuff (using Ollama and Whisper)
- Host: GitHub
- URL: https://github.com/lexiestleszek/talkr
- Owner: LexiestLeszek
- License: mit
- Created: 2024-02-17T11:52:39.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-02-19T10:49:04.000Z (over 1 year ago)
- Last Synced: 2025-01-13T21:44:13.979Z (5 months ago)
- Language: Python
- Size: 19.5 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Talkr: A Conversational Assistant
Talkr is a Python-based conversational assistant that listens for spoken commands, processes them, and provides a spoken response. It utilizes speech recognition and text-to-speech technologies to interact with users in a natural, conversational manner.
## Features
- **Speech Recognition**: Talkr uses the `speech_recognition` library to transcribe spoken commands into text.
- **Text-to-Speech**: The `pyttsx3` library is employed to convert text responses into spoken words.
- **Natural Language Processing**: The `whisper` library is used for speech-to-text transcription, and `ollama` for generating human-like responses.
## Getting Started
To get started with Talkr, you'll need to have Python installed on your system. Follow these steps:
1. Clone the repository:
```
git clone https://github.com/LexiestLeszek/talkr.git
```
2. Navigate to the project directory:
```
cd talkr
```
3. Install the required Python packages:
```
pip install -r requirements.txt
```
4. Run the main script:
```
python talkr.py
```
## Usage
Once the script is running, Talkr will listen for commands and respond accordingly. You can interact with it by speaking commands and it will provide responses.
## Contributing
Contributions to Talkr are welcome! If you have a feature request, bug report, or want to contribute code, please feel free to open an issue or submit a pull request.
## License
This project is licensed under the MIT License. See the `LICENSE` file for details.
## Contact
If you have any questions or suggestions, please open an issue.
## Acknowledgements
- The `speech_recognition` library for enabling speech recognition.
- The `pyttsx3` library for text-to-speech capabilities.
- The `whisper` and `ollama` libraries for natural language processing and response generation.