Skip to content

Latest commit

 

History

History
108 lines (80 loc) · 5.83 KB

README.md

File metadata and controls

108 lines (80 loc) · 5.83 KB
Raw

Introduction

The Transcribe solution is a multi-component application designed to transcribe audio files. It consists of a frontend, an API, a transcription worker, and uses external Speech to Text containers which are distributed by Microsoft.

Purpose

The purpose of this solution is to provide an automated way to transcribe audio files into text in a secure and scalable way. The solution is designed to be deployed as standalone containers or in a container orchestrator such as Kubernetes.

Target Audience

This README is intended for developers, contributors and users who want to understand, build, run, host and contribute to the Transcribe solution.

Documentation

Solution Components

  • Frontend: The frontend of the Transcribe solution is a React application located in the reactapp directory.
  • API: The API is a C# service located in the API directory.
  • Transcription Worker: The Transcription Worker is a service written in PowerShell that handles the transcription process. It's located in the TranscriptionWorker directory.
  • Speech to Text Containers: The solution uses external Speech to Text containers. More information can be found here.

Authentication

The application uses OpenID Connect for authentication of end users on the Microsoft identity platform (Entra ID).

For creation and configuration of App Registrations for the Frontend and API, see the following document:
Transcribe - Authentication setup_v1.0.pdf

Authorization for specific users to access the application can be configured in Entra ID under Enterprice application (frontend app registration). Currently the there is no concept of roles in the application and a logged in user can only see files/transcriptions uploaded by the user.

Building the Solution

Requirements

  • Docker
  • .NET SDK
  • Node.js and npm

Build Steps

  1. Build the API: dotnet build API
  2. Build the Frontend: npm install && npm run build in the reactapp directory.
  3. Build the Transcription Worker: dotnet build TranscriptionWorker
  4. Build the Docker images: docker-compose build

Development and Contribution Guidelines

How to Contribute

Please read the CONTRIBUTING.md for details on how to contribute.

How to Prepare the code for local development

  1. Create the file my_secrets.txt with content "APIKEY=mySecretAzureKey" in the project root.
  2. Make sure you have a developer certificate for the API application. This is done automatically for you when building using Visual studio. To do this manually from command line: dotnet dev-certs https -ep $APPDATA/ASP.NET/Https/API.pfx -p somepassword
  3. Make sure to register the certificate password as a secret. This is done automatically for you when building using Visual studio. To do this manually from command line: user-secrets set "Kestrel:Certificates:Development:Password" "somepassword" --project API
  4. Make sure to have ADFS authentication properly configured (see separate guide)
  5. Make sure you have a developer certificate for the reactapp application, follow vite guidelines.
  6. Review the settings in docker-compose.yml and docker-compose.override.yml and adjust if required.

How to Run the Solution Locally

  1. Build and start the services: docker compose -f docker-compose.yml -f .\docker-compose.override.yml up --build
  2. Access the frontend at http://localhost:4173
  3. Access the Backend at https://localhost:44331/swagger/index.html

Deploying the Solution

Requirements

  • An environment which can run Docker containers. This can be a local machine, a VM, or a container orchestrator such as Kubernetes.

For more details, please refer to the Infrastructure Design and Kubernetes Design documents.

Deployment Steps

Infrastructure

Each alternative deployment method has its own README file. Please refer to the README file for the deployment method you want to use.

Application Deployment

  1. Build the Docker images: docker-compose build
  2. Push the Docker images to a container registry.
  3. Update the image tags in the orchestrator deployment configuration.*
  4. Deploy the application.

* Follow the instructions in the README file for the deployment method you are using.

Updating the Solution

When new versions of the Speech to Text containers are released, update the image tag in the Kubernetes deployment configuration and redeploy the application.