Recent advancements in automatic speech recognition (ASR) and natural language processing (NLP) have made it feasible to efficiently summarize audio data, which is crucial given the vast amounts of audio content generated daily by companies. The process involves using OpenAI's Whisper ASR for transcription and GPT-3.5 for generating concise summaries. Whisper ASR is an open-source model introduced by OpenAI in 2022, capable of transcribing and translating audio data. It utilizes a sequence-to-sequence learning approach, employing encoder-decoder structures like Long Short Term Memory (LSTM) networks or Convolutional Neural Networks (CNN). For enterprise needs, Gladia offers an optimized version of Whisper ASR, addressing limitations such as hallucinations and long inference times. GPT-3.5, a large language model developed by OpenAI, excels in both extractive and abstractive summarization methods, providing accurate and concise summaries from transcriptions. The tutorial explains how to build an API using FastAPI for a seamless workflow that transcribes audio using Whisper or Gladia and then summarizes it with GPT-3.5. This integration enhances productivity by automating the extraction of key information from audio data.