techiaith-tts

---

Contents

Getting started:

• Installation

  • Installing with "pip"

  • Installing from source

• Overview

• Changelog

  • Unreleased

Installation

Installing with "pip"

pip install git+https://github.com/techiaith/techiaith-tts

Installing from source

git clone https://github.com/techiaith/techiaith-tts.git
cd techiaith-tts
conda create -n techiaith-tts python=[3.9 <= x.x => 3.12]
conda activate techiaith-tts
pip install -r requirements.txt
pip install -e .

Overview

The normaliser function parse_text() takes a string as an input and expands symbols to their string representation.

The code snippet below, from src/demo.py, runs an infinite loop taking user input and returning the full text representation of the input.

Usage

from techiaith.tts.testun.normaliser import parse_text

while True:
  text = input("\nprompt: ")
  clean_text = parse_text(text)
  print("\nresult:", clean_text)
  if clean_text == "exit":
    break

Output

$ python normaliser_test.py

prompt: Y dyddiad heddiw ydi'r 1af o Ionawr 2023
 
result: Y dyddiad heddiw ydi'r cyntaf o Ionawr dwy fil dau ddeg tri

Team

techiaith-tts is developed and maintained by the Uned Technolegau Iaith (UTI) https://techiaith.cymru/ team, backed by Prifysgol Bangor (UTI) https://bangor.ac.uk/. A self-funded research unit that develops technologies for the Welsh language. To learn more about who specifically contributed to this codebase, see our contributors page.

License

techiaith-tts is licensed under MIT License. A full copy of the license can be found on GitHub.

Indices and tables

• Index

• Module Index