parla-clean 0.1.0

Deterministic post-processing for Brazilian Portuguese speech transcription: filler removal, vocabulary variant correction, ASR deduplication
Documentation
1
2
3
4
5
6
7
8
9
10
11
12
//! Post-processing determinístico para transcrição pt-BR.
//!
//! # Uso
//! ```
//! let vocab = vec!["GitHub".to_string(), "Sam Altman".to_string()];
//! let clean = parla_clean::clean_text("tipo, o sematlman usa github né?", &vocab);
//! assert_eq!(clean, "O Sam Altman usa GitHub?");
//! ```

pub mod clean;

pub use clean::{clean_text, clean_text_with, CleanConfig, CleanResult, KNOWN_VARIANTS};