Voice-pathology analysis based on automatic topology generation of glottal source HMM

• Project/Area Number: 26330216
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: National Institute of Advanced Industrial Science and Technology
• Principal Investigator: SASOU AKIRA 国立研究開発法人産業技術総合研究所, 知能システム研究部門, 主任研究員 (50318169)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Project Status: Completed (Fiscal Year 2016)
• Budget Amount: ¥4,810,000 (Direct Cost: ¥3,700,000、Indirect Cost: ¥1,110,000); Fiscal Year 2016: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000); Fiscal Year 2015: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000); Fiscal Year 2014: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
• Keywords: 音声分析 / 声帯音源 / AR-HMM / 病的音声 / 声帯疾患 / 嗄声 / 聴覚印象評価 / 喉頭がん / GRBAS尺度 / 音声合成 / 食道発声音声 / 声質改善

Outline of Final Research Achievements

Voice-pathology detection from a subject's voice is a promising technology for the pre-diagnosis of larynx diseases. Glottal source estimation in particular plays a very important role in voice-pathology analysis. To more accurately estimate the spectral envelope and glottal source of the pathology voice, we propose a method that can automatically generate the topology of the Glottal Source Hidden Markov Model, as well as estimate the Auto-Regressive (AR)-HMM parameter by combining the AR-HMM parameter estimation method and the Minimum Description Length-based Successive State Splitting algorithm.

Research Products

• [Presentation] An evaluation of the automatically topology-generated auto-regressive hidden Markov model with regard to an esophageal voice enhancement task (2016)
  • Author(s): 佐宗晃
  • Organizer: 5th joint meeting of the acoustical society of America and acoustical society of Japan
  • Place of Presentation: Hilton Hawaiian Village Waikiki Beach Resort Honolulu, Hawaii, America
  • Int'l Joint Research
• [Presentation] Voice-pathology analysis based on AR-HMM (2016)
  • Author(s): 佐宗晃
  • Organizer: Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC2016
  • Place of Presentation: Ramada Plaza Jeju, Korea
  • Int'l Joint Research
• [Presentation] 駆動源ＨＭＭの出力分布に基づいた病的音声識別 (2015)
  • Author(s): 佐宗晃
  • Organizer: 日本音響学会2015年春季研究発表会
  • Place of Presentation: 中央大学後楽園キャンパス（東京都）
  • Year and Date: 2015-03-16
• [Presentation] Accuracy Evaluation of Esophageal Voice Analysis Based on Automatic Topology Generated-Voicing Source HMM (2014)
  • Author(s): 佐宗晃
  • Organizer: 15th Annual Conference of the International Speech Communication Association, Interspeech 2014
  • Place of Presentation: シンガポール
  • Year and Date: 2014-09-16