Automatic Speech Recognition — Capstone Project

1 minute read

An end-to-end Automatic Speech Recognition (ASR) system that accurately transcribes spoken language into written text across diverse domains and accents. The system is served via a FastAPI backend, presented through a Streamlit interface, and deployed on Azure Kubernetes Service (AKS).

App screenshot

Key Concepts

  • Automatic Speech Recognition (ASR) — converting spoken audio into written text with high accuracy across varied accents and subject domains
  • Kaldi — industry-standard open source toolkit for building speech recognition pipelines
  • Vosk API — lightweight offline speech recognition library that wraps Kaldi models for easy integration
  • FastAPI — high-performance Python web framework used to expose the ASR model as a REST API
  • Streamlit — Python-based UI framework for building and serving the interactive front end
  • Azure AKS — managed Kubernetes service used to deploy and scale the containerised application

Architecture

Audio Input (Streamlit UI)
        ↓
  FastAPI Backend
        ↓
  Vosk / Kaldi ASR Model
        ↓
  Transcribed Text Output
        ↓
  Deployed on Azure AKS

Prerequisites

Install the required dependencies:

pip install -r requirements.txt

Running the App

Start the FastAPI backend:

uvicorn main:app --reload

Launch the Streamlit UI:

streamlit run app.py

Repository

github.com/uday160386/asr-capstone-project

Presentation

You May Also Enjoy

Configuring Wifi in ESP32 WORM using code

1 minute read

Published:

Recently, I have been delving into a specific use case that involves consuming a voice REST endpoint using the ESP32 microcontroller. This task requires not only utilizing the capabilities of the ESP32 but also ensuring that the device is connected to a Wi-Fi network for seamless communication with the endpoint.

Data mocking using Faker

2 minute read

Published:

Ideally, test data is of priority and the project teams always face an issue in getting the relevant and realistic test data for pre-production activities. More issues(refresh of data; data manipulations etc.,) arise, when programs consume data from a shared environment. Sometimes, requirements of data varies and a new set of data should be replicated through external tools and technologies. Many commercial data mocking/stubbing tools are available in the market, but as a open source lover, I recommend using Faker library.