專利授權區國立清華大學國際產學營運總中心 Operations Center for Industry Collaboration

搜尋專利授權區

關鍵字

» 新增關鍵字

選單

專利授權區

專利授權區
專利名稱(中)	語音情緒辨識的方法、模型訓練方法與電腦程式產品
專利名稱(英)	METHOD, MODEL TRAINING METHOD AND COMPUTER PROGRAM PRODUCT FOR SPEECH EMOTION RECOGNITION
專利家族	中華民國：I859906
專利權人	國立清華大學 100.00%
發明人	簡婉軒,李祈均
技術領域	資訊工程,電子電機

專利摘要(中)
本發明提出一種語音情緒辨識的計算機實施方法，所述方法基於語音資料以及無偏見的語音情緒辨識模型，生成語音資料所對應的情緒預測結果。無偏見的語音情緒辨識模型是藉由根據已知偏見的有標籤訓練集以及損失函數訓練公平約束對抗網路來被訓練。公平約束對抗網路包括用以進行偏見分類的領域分類器以及所述語音情緒辨識模型，並且訓練公平約束對抗網路所使用的損失函數正相關於瓦瑟斯坦距離損失。

專利摘要(中)

本發明提出一種語音情緒辨識的計算機實施方法，所述方法基於語音資料以及無偏見的語音情緒辨識模型，生成語音資料所對應的情緒預測結果。無偏見的語音情緒辨識模型是藉由根據已知偏見的有標籤訓練集以及損失函數訓練公平約束對抗網路來被訓練。公平約束對抗網路包括用以進行偏見分類的領域分類器以及所述語音情緒辨識模型，並且訓練公平約束對抗網路所使用的損失函數正相關於瓦瑟斯坦距離損失。

專利摘要(英)
A computer-implemented method for speech emotion recognition is provided. According to this computer-implemented method, an emotion prediction corresponding to speech data is generated based on the speech data and a speech emotion recognition model without bias. The speech emotion recognition model without bias is trained by training a fairness-constrained adversarial network based on a labeled training set with known bias and a loss function. The fairness-constrained adversarial network includes a domain classifier for bias classification and the speech emotion recognition model. The loss function used for training the fairness-constrained adversarial network is positively related to a Wasserstein distance (WD) loss.

專利摘要(英)

A computer-implemented method for speech emotion recognition is provided. According to this computer-implemented method, an emotion prediction corresponding to speech data is generated based on the speech data and a speech emotion recognition model without bias. The speech emotion recognition model without bias is trained by training a fairness-constrained adversarial network based on a labeled training set with known bias and a loss function. The fairness-constrained adversarial network includes a domain classifier for bias classification and the speech emotion recognition model. The loss function used for training the fairness-constrained adversarial network is positively related to a Wasserstein distance (WD) loss.

聯絡資訊
承辦人姓名	李曉琪
承辦人電話	03-5715131 #31061
承辦人Email	hsiaochi@mx.nthu.edu.tw