"ASR to Graph Knowledge"

Read full text Summarise with VIRAT
Authors: Emmanuel Ukekwe
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

"Automatic speech recognition researches are gaining a lot of attention nowadays. However, getting requisite data sets with corresponding transcripts for ground truth comparison is proving difficult. This data set contains requisite links and information to audio lecture recordings and corresponding transcripts. It also contains already computed P_k, WindowDiff, ARI, SemenaticCohesion, Embedding Precision, Embedding Recall, Embedding F1, TFIDF_Precision, TFIDF_Recall, TFIDF_F1, Whisper_WER, Whisper_CER, and GroundTruth_WER. The essence of the dataset is to make it easier for researchers in speech recognition to have an already made data to work with. The already computed metrics is an added advantage. The computations were done using python"

Fields of Study & Subjects

Speech Recognition and Synthesis Speech and Audio Processing Handwritten Text Recognition Techniques

Publication details

Journal / source
IEEE DataPort
Publication date
30 March 2026
Language
EN
Document type
Dataset