• 履歴一覧
• 差分 を表示
• 現在との差分 を表示
• ソース を表示
• HOME/DEMONSTRATION/SPEAKER ADAPTATION へ行く。
  • 1 (2008-09-02 (火) 14:42:18)
  • 2 (2008-09-02 (火) 14:47:24)
  • 3 (2008-09-03 (水) 06:58:51)
  • 4 (2008-09-03 (水) 11:33:48)

DEMONSTRATION - Speaker adaptation

In the speech synthesis system, a large amount of trainin data is required.
However, the cost for recoding speech is unaffordable.
Speaker adaptation is a technique for control style with a small amount of adaptation data.

• Speech of the target speaker
  "nazejibuNbakarikoNnameniaunodarou"
  「なぜ自分ばかりこんな目にあうのだろう．」
  • target speech

• Synthesis speech with speaker adaptation
  "nazejibuNbakarikoNnameniaunodarou"
  「なぜ自分ばかりこんな目にあうのだろう．」
  • without adaptation
    synthesis speech
  • with adaptation using one sentence
    synthesis speech
  • with adaptation using three sentences
    synthesis speech
  • with adaptation using five sentences
    synthesis speech
  • with adaptation using seven sentences
    synthesis speech

We could perform controling style with only a few adaptation data.
By using this technique, we will be able to construct the speech synthesis sytem which have higer quality with a small amount of training data.