Bangla ASR Improvement: Novel Corpus and Analysis for Disfluency Detection
Analysis
This research addresses a critical challenge in Automatic Speech Recognition (ASR) for the Bangla language, focusing on differentiating between repetition disfluencies and morphological reduplication. The creation of a novel corpus and benchmarking analysis is a significant contribution to the field.
Key Takeaways
Reference / Citation
View Original"The research focuses on distinguishing repetition disfluency from morphological reduplication in Bangla ASR transcripts."