Incremental Structuring of Speech Data Considering Relationships with Readability of Transcription

• Project/Area Number: 16K16119
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Intelligent informatics
• Research Institution: National Institute of Technology, Toyota College
• Principal Investigator: MURATA Masaki 豊田工業高等専門学校, 情報工学科, 助教 (30707807)
• Research Collaborator: SHIGEHISA Naoki
• Project Period (FY): 2016-04-01 – 2018-03-31
• Project Status: Completed (Fiscal Year 2017)
• Budget Amount: ¥3,900,000 (Direct Cost: ¥3,000,000、Indirect Cost: ¥900,000); Fiscal Year 2017: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000); Fiscal Year 2016: ¥2,600,000 (Direct Cost: ¥2,000,000、Indirect Cost: ¥600,000)
• Keywords: 音声言語 / 字幕生成 / 構造化 / 漸進的解析 / 自然言語処理

Outline of Final Research Achievements

I have promoted the recording of research data and the construction of a part of structured data. I have constructed examples of parsing results and considered annotation guidelines for existing discourse data. Also, I have recorded speech and image data of five classes. A part of speech data was transcribed into texts.
I have developed a prototype of a captioning system. The system displays the result of speech recognition on PowerPoint by use of a VSTO add-in for PowerPoint. A subjective evaluation shows a certain evaluation for display time and readability of caption.