yoruba-tts
v1.0.4
Published
A wrapper for a Yoruba TTS Model
Downloads
1
Readme
Yoruba Text-to-Speech Package
A simple Node.js package to generate speech from text using the Meta's mms-tts-yor model on HuggingFace API. It is also possible to make use of another model which might be more advance than this one.
Installation
npm install yoruba-tts
Usage
Remember to get your API token from HuggingFace and put in the .env file in the root of your project.
TTS_MODEL_URL=https://api-inference.huggingface.co/models/facebook/mms-tts-yor
HUGGINGFACE_API_KEY=
import { generateSpeech } from 'yoruba-tts'
const text = "Mo fẹ́ràn Èdè Yorùbá";
const outputPath = "./output.flac";
generateSpeech(text, outputPath);
The output will be a flac file in the location specified.
Sample Audio
Sample Audio: feran.flac
Credit
The credit for the model belongs to Vineel Pratap et al. from Meta AI.
@article{pratap2023mms,
title={Scaling Speech Technology to 1,000+ Languages},
author={Vineel Pratap and Andros Tjandra and Bowen Shi and Paden Tomasello and Arun Babu and Sayani Kundu and Ali Elkahky and Zhaoheng Ni and Apoorv Vyas and Maryam Fazel-Zarandi and Alexei Baevski and Yossi Adi and Xiaohui Zhang and Wei-Ning Hsu and Alexis Conneau and Michael Auli},
journal={arXiv},
year={2023}
}
Contact
The author can be contacted for feedback at Surajudeen Akande