AssemblyAI Launches Starter App for Encore to Simplify Speech AI Development

Tony Kim  Jun 20, 2024 13:20  UTC 05:20

0 Min Read

AssemblyAI has unveiled a new starter app for Encore, aiming to simplify the process of developing and deploying Speech AI applications. This new tool is designed for Go developers and offers a comprehensive solution to transcribe audio files and analyze conversations, according to AssemblyAI.

What is Encore?

Encore is a development platform that facilitates the creation and management of APIs and event-driven systems. With a generous free tier, it is particularly appealing for developers working on prototypes or hobby projects. The platform streamlines the process of getting a Speech AI application up and running, making it easy to share with team members or friends.

Introducing the AssemblyAI Starter App

The newly introduced assemblyai-starter app allows developers to bootstrap a Speech AI application using a single command. To get started, users need to install the Encore CLI:

  • macOS: brew install encoredev/tap/encore
  • Windows: iwr https://encore.dev/install.ps1 | iex
  • Linux: curl -L https://encore.dev/install.sh | bash

Once installed, developers can create an Encore app from the AssemblyAI starter template and configure it using their AssemblyAI API key and webhook secret. The app can then be deployed with a simple Git push command.

How to Use the App

The AssemblyAI starter app features a user-friendly interface to upload local audio files for transcription. The app supports Speaker Diarization, allowing users to identify who spoke and when. Transcriptions are managed through a straightforward UI, where users can upload files, monitor transcription status, and view completed transcriptions.

App Architecture Overview

The app architecture comprises three main components:

  • A frontend built with React and Tailwind CSS
  • A backend developed using Go and the AssemblyAI Go SDK
  • A PostgreSQL database to store completed transcripts

The workflow involves uploading audio files through the frontend, which are then submitted to AssemblyAI for transcription via the backend. Once the transcription is complete, the data is stored in the database, and the frontend retrieves and displays the transcribed conversation.

To avoid constantly polling AssemblyAI for updates, the app uses webhooks to notify the backend when a transcription is ready.

Developing Locally

For local development, Docker is required to replicate the cloud environment. Developers can run the application locally with the encore run command, which sets up a Docker-based environment mirroring the cloud setup. Any backend changes will build and restart the service automatically, while frontend changes require a rebuild using npm run build.

Summary

The AssemblyAI starter app for Encore offers a robust starting point for Go developers interested in exploring Speech AI. With easy setup and comprehensive features, it simplifies the process of building, deploying, and managing Speech AI applications.



Read More