SN-WER normalizes scripts via transliteration before WER computation, reducing apparent model gaps by up to 12% on FLEURS data for five Indic languages while remaining sensitive to real lexical errors.
Lenient Evaluation of J apanese Speech Recognition: Modeling Naturally Occurring Spelling Inconsistency
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
SN-WER: Script-Normalized WER for Multi-Script Indic ASR Evaluation
SN-WER normalizes scripts via transliteration before WER computation, reducing apparent model gaps by up to 12% on FLEURS data for five Indic languages while remaining sensitive to real lexical errors.