Courtroom Transcription: A Deep Learning Approach to Legal Terminology and Speaker Identification

機構典藏 > College of Engineering > Graduate Institute & Department of Computer Science and Information Engineering > Proceeding > Item 987654321/126360

Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/126360

Title:	Courtroom Transcription: A Deep Learning Approach to Legal Terminology and Speaker Identification
Authors:	Jhang, Syu-Jhih;Chang, Hsiang-Chuan;Zhang, Qiao-yun;Chang, Chih-Yung
Keywords:	Speech Recognition;Legal Terminology;Deep Learning;Role Identification;Error Correction
Date:	2024-07-09
Issue Date:	2024-10-07 12:06:04 (UTC+8)
Abstract:	This study develops a deep learning with natural language processing. It focuses on overcoming the limitations of current speech recognition tools, which struggle with legal terminology and identifying different courtroom speakers. By combining advanced audio processing, role identification, and error correction techniques, including a Bert-based model and an N-gram model, the research aims to automate the transcription process more efficiently. This method not only promises to enhance the accuracy of capturing court proceedings but also aims to revolutionize the transcription practices by reducing manual effort and increasing the reliability of legal documents.
Appears in Collections:	[Graduate Institute & Department of Computer Science and Information Engineering] Proceeding

Files in This Item:

There are no files associated with this item.