GenAI: Function Calling, Whisper, LangChain (UPDATED 2024)

Explore OpenAI's language models, learn how to authenticate and create an account, and start coding with Python or Node.js SDKs, API references, and quick-start examples.

Download the starter project for factual answering, run a command-line interface to ask AI factual questions, and configure OpenAI API keys and language models to generate completions.

Set up the quickstart environment for OpenAI by installing Python, creating a virtual environment, installing the OpenAI library, and configuring an API key to run the main.py interface.

Configure your project with an API key to interact with the language model, set up environment variables, and run a test prompt using Python with the legacy completions API.

Send API requests to a language model, convert user input to strings, print the model’s response, and format output with Colorama while tuning tokens and temperature for stable answers.

Discover how GPT models enable human-like conversations in ChatGPT and AI chatbots. Learn to set up a Node.js project, authenticate with OpenAI, and craft prompts for chat completions.

Install and run the starter project for a chat bot with express, cors, and OpenAI. Build endpoints and a client UI, using environment variables for the OpenAI key.

Define a system message and a prompt to guide a language model’s behavior, including personas like a helpful assistant, a professional writer, or a senior UX designer.

Pass user input as content in API requests to the language model, manage conversation history as a list of messages, and enable looping conversations for ongoing chat.

Execute final steps to create a restful API and a React-based client UI for a chat bot, structuring messages with roles and wrapping calls in try-catch.

Set up a node server with express to power a full-stack chat bot, building a RESTful API with an index.js entry point and configuring npm start on port 5000.

Define prompts and send requests to a language model via a post route in a rest api, configure api keys, system messages, and json handling, then test with curl.

Test a RESTful API with postman by sending a post request to the ask route on port 5000 and validate chatbot's response, planning a UI for AI chat bots.

Build a chatbot user interface using react, axios, and bootstrap to create a responsive, user friendly client app and integrate a chatbot feature.

Set up a react chat bot interface and manage input and message state. Send user messages via an Axios API request to a language model using a restful backend.

Define requests with axios by sending a post to the RESTful API, passing a messages array in the body, handling responses, and configuring CORS for localhost cross-origin access.

Install cross-origin resource sharing middleware at server root, add it as middleware, and update package.json to enable cross-origin requests; run npm start to test and view dialogue in the app.

Demonstrates rendering and formatting chat messages in a client app with system, user, and assistant roles, and introduces function calling to integrate external data from OpenAI models.

Learn to enable a language model to call external APIs via function calling, using a chat endpoint and json prompts to fetch current weather data.

Kick off a starter project by defining a function and letting the language model auto call it to fetch weather data from an external API using user messages.

Learn to fetch current weather from openweathermap.org using latitude, longitude, and an API key, including geocoding to coordinates and converting Kelvin to Celsius or Fahrenheit.

Learn how to test function calling with extended responses, decide when to return regular vs extended messages, wrap logic in try-catch, and filter messages for clean API-driven conversations.

Build a full-stack chatbot that uses function calling to fetch current weather data from an external API and display it within a natural chat conversation.

Create a starter rest API with FastAPI and Python to download audio, transcribe it to text using Whisper, and run a local uvicorn server with CORS and environment setup.

Learn to download an audio file from a URL to the file system, validate the download with a test request, and prepare for transcription with the whisper model.

Learn how to convert audio to text by loading and using a whisper model to transcribe an audio file, handle missing files, and return the transcript via an API response.

Transcribe podcast audio with whisper, extract and return transcripts, and test the workflow with postman and a graphical user interface.

Learn to build a speech-to-text application using LangChain and the OpenAI Whisper API to convert audio into transcripts and generate summaries.

Install dependencies, start the project with npm start, and verify downloading an mp3 from a URL, then transcribe with whisper and summarize the text.

Set up the OpenAI whisper audio model by loading an audio file, creating a loader, and transcribing to documents, then summarize the podcast transcription.

Build a transcription application by running a summarization chain over transcripts using a wrapper OpenAI model, loader, and text splitter, converting transcripts into documents and generating concise summaries.

Run tests with postman to validate endpoints, download an MP3, and generate transcripts and summaries for audio files, demonstrating end-to-end transcription for podcast workflows.

Connect and run a React client app, install dependencies with npm i, start the server, and perform transcription and summarization via Axios, displaying the transcript and summary.

Complete the final steps to transcribe an audio URL via a post request, show a spinner during processing, handle errors, and update transcript and summary.