← Back to Patent Explorer
JPH064097A Expired / Lapsed JP ⏱️ 0.0 years remaining

Speaker recognition method

Assignee: Seiko Epson Corp
•
Filing: 1992-06-18
•
Granted: —
•
Expires: 2012-06-18
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract Translated from Japanese (57)【要約】 【目的】 本発明はより高性能の話者認識を実現する事 を目的とする。 【構成】 本発明は内部状態値記憶部を持ち、かつその 内部状態値を内部状態値そのものと外部入力値の両方の 値により更新するニューラルネットワークと、音声特徴 抽出手段により構成される話者認識方法である。このニ ューラルネットワークにより、入力の復元、予測等の精 度により話者認識をおこなう。 【効果】 従来例に比較して非常に高速な、かつ正確な 認識を行う事ができる。また、それを実行するためのハ ードウェアも非常に小さくする事ができる。 (57) [Summary] [Object] An object of the present invention is to realize higher-performance speaker recognition. According to the present invention, there is an internal state value storage unit, and a speaker recognition including a neural network that updates the internal state value with both the internal state value itself and an external input value, and a voice feature extraction means. Is the way. With this neural network, speaker recognition is performed with the accuracy of input restoration and prediction. [Effect] It is possible to perform recognition at extremely high speed and accuracy as compared with the conventional example. Also, the hardware to implement it can be very small.

Inventors

  • 👤 Mitsuhiro Inazumi
  • 👤 満広 稲積

Bibliographic Details

Application Number —
Jurisdiction JP
Data Source google_patents
Last Synchronized 2026-11-08 13:17
CPC Classifications
—

Claims Analysis

Total Claims: 0
No claims parsed yet. Click Sync Now above to re-index claims from Google Patents.

References Cited (Backward Citations)

No citation records indexed.

USPTO Annuity & Maintenance Windows

Maintenance schedule is calculated for granted US utility patents.

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag