How cabird/quickscribe works
Main Components
• Frontend – Represented by the “vite-frontend” directory:
• Uses Vite, TypeScript, and React (or a similar framework based on file names such as App.tsx, Router.tsx, and various UI components).
• Contains API calls (see the “api” folder) that interact with the backend.
• Backend – Represented by the “backend” directory:
• Contains the primary application code (app.py, manage.py) and routes (in the “routes” folder) that expose REST endpoints.
• Includes specialized modules such as “audiostream_api” for audio streaming operations and “azure_speech” which integrates with Microsoft Azure’s Speech services via its autogenerated Python client.
• Contains various API endpoints: audiostream, transcription routes (az_transcription_routes.py), etc.
• Database and Storage – Evidenced by:
• “db_handlers” directory which likely manages interactions with the underlying databases (Cosmos DB hints in “cosmos_schema” and file names like recording_handler.py, transcription_handler.py).
• The “chunk_storage” files in audiostream_api indicate that audio data might be split into chunks for processing or storage.
• External Services & Third-Party Integrations:
• Azure Speech Services: The “azure_speech” folder (with its Python client, swagger_client, and extensive documentation) signals integration with Microsoft Azure’s cognitive services.
• Infrastructure – Represented by the “infrastructure/terraform” directory:
• Contains scripts (e.g., create_resources.sh) for provisioning cloud resources. This suggests that your deployment may involve automated cloud resource management.
• Shared Models:
Relationships and Interactions
• Frontend to Backend:
• Backend Internal Modules:
• Within the backend, route handlers in “routes” delegate operations to service layers defined in “audiostream_api,” “db_handlers,” and custom utilities like “auth.py”, “llms.py”, and “util.py.”
• Integration with Azure:
• Database Operations:
• Infrastructure:
• “Terraform/Infrastructure” as a separate component that provisions or manages cloud resources, which in turn support the backend’s hosting or the database’s deployment.
Architectural Patterns and Design Principles
• Layered Architecture:
• Separate the presentation layer (frontend), business logic (backend routes, APIs, and service modules), and data persistence (database handlers).
• Separation of Concerns:
• RESTful Architecture: