A Short-Time Speech Analysis Method with Mapping Using the Fejer Kernel (Special Section on Signal Processing and System Theory)
スポンサーリンク
概要
- 論文の詳細を見る
We discuss estimation error as a basic problem in formant estimation in the analysis of speech of very short-time duration in the glottal closure of the vowel. We also show in our simulation that good estimation of the first formant is almost impossible with the ordinary method using a waveform cutting. We propose a new method in which the cut waveform, as a discontinuous function of finite time, is mapped to a continuous function defined in the whole time domain; and we show that using this method, the estimation accuracy for low frequency formants can be greatly improved.
- 社団法人電子情報通信学会の論文
- 1994-05-25
著者
-
Nagai Nobuo
Research Institute for Electronic Science, Hokkaido University
-
Miki Nobuhiro
Research Institute Of Applied Electricity Hokkaido University
-
Nagai Nobuo
Research Institute For Electronic Science Hokkaido University
-
Takemura Kenji
Research Institute for Electronic Science, Hokkaido University
-
Takemura Kenji
Research Institute For Electronic Science Hokkaido University
関連論文
- An ARMA Order Selection Method with Fuzzy Theorem (Special Section of Papers Selected from JTC-CSCC'93)
- Relation between RLS and ARMA Lattice Filter Realization Algorithm and Its Application (Special Section on Signal Processing and System Theory)
- Measurement of acoustic reflection characteristics of the human cheek
- Measurement of radiation characteristics using replicas of the lips
- A Restatement on Applications of Electrical Considerations for One-Dimentional Wave Phenomena (Special Section on Signal Processing and System Theory)
- A Short-Time Speech Analysis Method with Mapping Using the Fejer Kernel (Special Section on Signal Processing and System Theory)