This repository contains personal projects exploring Large Language Model (LLM) capabilities, primarily focusing on OpenAI's API for text-to-speech and chat completions. It targets developers interested in practical implementations of LLM features like audio generation and conversational AI, offering code examples and demonstrations of streaming audio and text.

How It Works

The projects leverage OpenAI's API for core LLM functionalities. Button.py demonstrates streaming audio from OpenAI's Text-to-Speech (TTS) endpoint, providing a basic UI for input and output. narrator.ipynb processes video files to generate narrated content. streamed_text_plus_streamed_audio.py combines streaming text responses from GPT-3.5-turbo with chunked, streamed audio output from TTS, offering a more interactive experience.

Quick Start & Requirements

• Install dependencies via pip install -r requirements.txt.
• Requires an OpenAI API key.
• Demo video available via link in README.

Highlighted Details

• Provides a working example for OpenAI's streaming TTS, which is noted as lacking in official documentation.
• Demonstrates chunking text responses for incremental TTS processing.
• Includes a script for generating narration from video files.

Maintenance & Community

This is a personal repository; no specific community or maintenance details are provided. The author mentions using "Double.bot" as their AI copilot.

Licensing & Compatibility

The repository does not explicitly state a license.

Limitations & Caveats

The projects are personal experiments and may not be production-ready. The narrator.ipynb script requires tweaking system prompts for optimal results. OpenAI API usage incurs costs.