saarus72/text_normalization

T5-based (russian) text normalization

30
/ 100
Emerging

This project helps convert raw, informal Russian text into a more standardized format suitable for text-to-speech (TTS) systems. It takes everyday Russian sentences that might include numbers, abbreviations, or mixed-language terms (like 'iphone 10X') and converts them into their fully written-out, grammatically correct Russian equivalents. Anyone working with Russian speech synthesis, such as content creators, linguists, or educators, would find this useful.

No commits in the last 6 months.

Use this if you need to reliably transform casual, written Russian text into a clean, spoken-word-ready format for speech synthesis applications.

Not ideal if your primary need is for inverse text normalization (converting spoken numbers back to digits), although it includes components that assist with that.

Russian speech synthesis text preprocessing natural language processing audio production linguistics
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 7 / 25

How are scores calculated?

Stars

26

Forks

2

Language

Jupyter Notebook

License

Apache-2.0

Last pushed

Jan 25, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/saarus72/text_normalization"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.