Seq2Seq English to IPA Phonetics Model API

This API exposes an English to International Phonetic Alphabet (IPA) conversion Tensorflow model. The model is a proof of concept for transforming English text into its corresponding IPA phonetic representation. The end goal of this project is to aid the generation of IPA phonetics dictionaries for various languages and dialects.

Model Repo

Sequence to Sequence English to IPA Phonetics Model

Prerequisites

Before using this API, ensure that you have installed:

Node.js
AWS CLI
AWS CDK
Docker
PNPM

If you have not installed these prerequisites, please refer to their official documentation for detailed installation instructions.

Quick Start

Clone this repository:

git clone https://github.com/OrderAndCh4oS/english-to-ipa-phonetics-ml-cdk
cd english-to-ipa-phonetics-ml-cdk

Install dependencies:

pnpm install

Deploy your stack via CDK:

pnpm cdk:deploy

After successful deployment, access the API through provided endpoint URL in the deployment output.
To remove the stack from AWS after use, run:

pnpm cdk:destroy

API Documentation

The Seq2Seq English to IPA Phonetics API provides an easy-to-use endpoint for converting written English text into its corresponding IPA phonetic representation.

POST /

Request:

Content-Type: application/json

Body:

{
    "text": "The quick brown fox jumps over the lazy dog"
}

Response:

Content-Type: application/json

Body:

{
    "ipa": "/θˈiː/ /kwˈɪk/ /bɹˈaʊn/ /fˈɒks/ /dʒˈʌmps/ /ˈəʊvɐ/ /θˈiː/ /lˈeɪzi/ /dˈɒɡ/"
}

License

This project is licensed under the terms of the MIT license.

Non-ML Language to IPA Repos

Phonetics Translator NextJS Phonetics Translator Scripts & Dictionaries