Skip to content

khanld/speechenhancement

Repository files navigation

SPEECH ENHANCEMENT FOR ASR USAGE

Table of contents

  1. Description
  2. Usage
  3. Web App
  4. API Calls

Description

This is a demo for FullSubNet Speech Enhancement for Vietnamese ASR. The Speech Enhancement model was trained on 1k3 hours of speech data with dynamic mixing (mix-on-the-fly).
Here, we found an approach to remedy the problem of SE when adapting as front-end to ASR (SE causes the degradation of ASR decoding performance on clean speech). The idea is simple but can work with any SE and ASR models. alt text

Usage

Download checkpoints

  1. First, download the model checkpoints from this link
  2. Move the downloaded folder to the root directory and rename it as "checkpoints" alt text
  3. Set up environment with either Docker or Manually

Docker

Docker build

docker build -t demo .

Docker run

Docker run demo

Manually

Install packages

pip install -r requirements.txt

Run Flask app

python app.py -c config.json

Web App

alt text

API Calls

Check example.py file for API usage