J.A.R.V.I.S is a "Voice-Only Assistant" refers to a type of digital assistant or artificial intelligence system that primarily interacts with users through voice commands and responses, without relying heavily on visual interfaces or screens. These assistants are designed to understand spoken language and provide information or perform tasks based on voice inputs.
This app was created as part of the Generative AI & App Dev course at Kasetsart University.
- Installation : follow the step in Installation
- Running : run by typing this code in your terminal, cmd or powershell.
python Jarvis.py
- Step 1: Speech Recognition:
- Listening to human voice, then recognizing those human voice and transforming it into a sequence of text
- Step 2: Text-to-Text (Sequence-to-Sequence) Generation:
- Sending that text (supposing to be a human conversation) as an input into a LLM, like OpenAI's GPT or any open-source LLM's (through APIs).
- Step 3. Speech Synthesis:
- The output from Step2 (the sequence of text output from the LLM) to be transformed into speech.