ファジィ目的関数最小化に基づく短時間教師なし話者適応
スポンサーリンク
概要
- 論文の詳細を見る
This paper presents an unsupervised speaker adaptation method from short utterances. The code spectra for the templates are adapted to those of an input speaker by interpolating the estimated speaker-difference vectors for given typical spectra. These difference vectors are estimated so as to minimize the fuzzy objective function for the adapted reference codebook under some constraints. The fuzziness (F) and constraint parameters are examined using SPLIT-based word recognition tests with 28-word vocabulary and reference patterns from a male speaker. Using 1. 8s training samples, the results show that the proposed method with F = 1. 5 gives a 9. 0% without adaptation to 97. 5% using 3. 6s training samples. Furthermore, a sequential adaptation scheme attains an average recognition rate of 97. 4% using test speech itself for adaptation.
- 社団法人日本音響学会の論文
著者
-
山下 泰樹
長野県工科短期大学校
-
松本 弘
Department of Electrical and Electronic Engineering,Faculty of Engineering,Shinshu University
-
山下 泰樹
Department of Electrical and Electronic Engineering,Faculty of Engineering,Shinshu University
-
山下 泰樹
Department Of Electrical And Electronic Engineering Faculty Of Engineering Shinshu University
-
松本 弘
Department Of Electrical And Electronic Engineering Faculty Of Engineering Shinshu University
関連論文
- 成人の読上げ音声における声質評価値と音響関連量の分析的検討
- 動的個人性に関する声質評価値と音響的関連量の検討
- 声質評価値と音響パラメータの間の分析的検討
- 声質評価値と音響パラメータとの関連の検討-第2報
- 声質評価値と音響パラメータとの関連の検討
- 音声の動的個人性に関する主観評価の試み
- ファジィ目的関数最小化に基づく短時間教師なし話者適応
- ベクトル量子化誤差の区分平均化による教師なし話者適応
- 歌声の心理的印象と音響特徴量との対応付けによる歌唱の習熟度評価に関する基礎的検討