This paper presents a simple sequence-to-sequence approach to restore standard orthography in raw, normalized speech transcripts, including insertion of punctuation marks, prediction of capitalization, restoration of numeric forms, formatting of dates and times, and other, fully data-driven adjustments. We further describe our method to generate synthetic parallel training data, and explore suitable performance metrics, which we align with human judgment through subjective MOS-like evaluations. Our models for English, Russian, and German have a word error rate of 6.36%, 4.88%, and 5.23%, respectively. We focus on simplicity and reproducibility, make our framework available under a BSD license, and share our base models for English and Russian.
Paper
Full text
Neural Text Denormalization for Speech Transcripts
Semantic Scholar · Computer Science · 2021
Abstract
This paper presents a simple sequence-to-sequence approach to restore standard orthography in raw, normalized speech transcripts, including insertion of punctuation marks, prediction of capitalization, restoration of numeric forms, formatting of dates and times, and other, fully data-driven adjustments. We further describe our method to generate synthetic parallel training data, and explore suitable performance metrics, which we align with human judgment through subjective MOS-like evaluations. Our models for English, Russian, and German have a word error rate of 6.36%, 4.88%, and 5.23%, respectively. We focus on simplicity and reproducibility, make our framework available under a BSD license, and share our base models for English and Russian.