Practical overview
When it fits
Use it when evaluating or adapting a voice-based RAG application that retrieves from Azure AI Search and generates spoken responses through Azure OpenAI.
Who it is for
- Developers exploring voice interfaces for retrieval-augmented generation.
- Teams adapting an Azure-hosted knowledge-base assistant to accept spoken questions.
- Builders who want a reference pattern for combining Azure AI Search with realtime audio generation.
Documented uses
- Ask a knowledge-base question using browser microphone input.
- Retrieve relevant documents from Azure AI Search and provide them to the realtime audio model as context.
- Play an audio response and show the search results used to generate it.
- Use the provided sample and infrastructure code as a starting point for an Azure Container Apps deployment.
Requirements & limits
The repository is public and identifies its license as MIT. The documented deployment provisions Azure resources that may incur costs.
- The retrieval and generation workflow depends on Azure AI Search and Azure OpenAI services.
- The browser microphone captures voice input, while browser audio capabilities play generated responses.
- Search results used for an answer are shown to the user.
- The repository provides Codespaces, VS Code Dev Containers and local setup paths; local setup lists Azure Developer CLI, Node.js, Python 3.11 or later and Git, with PowerShell for Windows.
- The repository includes a Dockerfile and infrastructure as code for Azure Container Apps deployment.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Prepare the local development environment
Set up the documented tools and clone the sample repository before proceeding to its deployment instructions.
Before you start
- Azure Developer CLI
- Node.js
- Python 3.11 or later, with Python and pip available on PATH for Windows setup scripts
- Git
- PowerShell on Windows
Steps
Install the listed tools for your environment.
Clone the repository with git clone https://github.com/Azure-Samples/aisearch-openai-rag-audio.
Proceed to the repository's deployment instructions.
Expected outcome
The sample source is available locally with the prerequisite development tools installed.
Guide limitations
- This setup recipe does not deploy or run the application.
- The application requires configured Azure AI Search and Azure OpenAI services for local use.
Provision and deploy the sample to Azure Container Apps
Use the documented Azure Developer CLI workflow to provision resources, deploy the code and configure integrated vectorization for sample data.
Before you start
- Azure account
- Azure Developer CLI
- A prepared development workspace, such as the documented Codespaces, VS Code Dev Container or local environment
Steps
Sign in with azd auth login; Codespaces users can use azd auth login --use-device-code if the standard command fails.
Create an environment with azd env new and enter a name for its resource group.
Optionally configure environment variables for existing services or voice choice, following the repository guidance.
Run azd up to provision resources, deploy the application and set up integrated vectorization for sample data.
Expected outcome
The documented workflow provisions Azure resources and deploys the application to Azure Container Apps.
Guide limitations
- The README warns that newly created resources can incur immediate costs, particularly Azure AI Search, including if provisioning is interrupted.
- The recipe describes the repository's deployment procedure, not a verified deployment or compatibility guarantee.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 9 October 2026
- Original website title
- Azure-Samples/aisearch-openai-rag-audio · Official homepage title · retained in its source language