This repository demonstrates how to use the Azure AI Services to transcribe audio files and summarize the results using Azure OpenAI.
The implementation is based on two functions:
- transcription-manager: that gets triggered when new audio files have been uploaded to a blob storage container. The function uses the Speech to text API to transcribe the audio file and stores the result in a Cosmos DB container.
- transcription-summarizer: that gets triggered when new transcriptions have been stored in the Cosmos DB container. The function uses the OpenAI API and Semantic Kernel to summarize the transcription and stores the result in a Cosmos DB container.
The high level architecture is shown in the following diagram:
The steps to deploy the whole solution are described in the following sections.