Zobacz książki o podobnej tematyce

Więcej informacji o Self-Learning Speaker Identification

• Dane szczegółowe
• Opis
• Ulubione w innej kategorii

Za ten zakup dostaniesz 331 punkty

Opis

Current speech recognition systems suffer from variation of voice §characteristics between speakers as they are usually based on speaker §independent speech models. In order to resolve this issue, adaptation §methods have been developed in many state-of-the-art systems. However, §information acquired over time is still lost whenever another speaker intermittently§uses the recognition system. This work therefore develops an integrated§approach for speech and speaker recognition in order to improve the§self-learning opportunities of the system. A speaker adaptation scheme§is introduced. It is suited for fast short-term and detailed long-term§adaptation. These adaptation profiles are then used for an efficient§speaker recognition system. The speaker identification enables the§speaker adaptation to track different speakers which results in an§optimal long-term adaptation.

Szczegóły książki

Kategoria Książki po angielsku Computing & information technology Computer science Artificial intelligence

• Pełny tytuł: Self-Learning Speaker Identification
• Podtytuł: A System for Enhanced Speech Recognition
• Autor: Tobias Herbig, Franz Gerl, Wolfgang Minker
• Język: Angielski
• Oprawa: Twarda
• Liczba stron: 172
• EAN: 9783642198984
• ISBN: 3642198988
• ID: 01657619
• Wydawca: Springer-Verlag Berlin and Heidelberg GmbH & Co. KG
• Waga: 970 g
• Wymiary: 235 × 155 × 20 mm
• Data wydania: 18. June 2011

Ulubione w innej kategorii

---

---

---