PERL: Pinyin Enhanced Rephrasing Language Model for Chinese ASR N-best Error Correction
Junhong Liang · 2024
Read PERL: Pinyin Enhanced Rephrasing Language Model for Chinese ASR N-best Error Correction by Junhong Liang (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-03230.txt.
Description
Junhong Liang and Bojun Zhang present a 2024 paper on fixing errors in Chinese speech recognition. Errors are often phonetic, since many characters share the same Pinyin sound spelling. A fixer must also keep the output length right while it picks among noisy candidate lists from the recognizer.
Their system, PERL, first predicts the target length and enforces it by a mask budget. It then blends meaning and Pinyin sound cues through gates set by the sense of each sentence. On a standard test set and a new domain benchmark, the system cut character error rates steeply, with low delay. Study of its choices shows when it leans on sound and when on sense.
Catalog record
| CREATOR | Junhong Liang |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 21 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-03230