arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2203.09148cs.SDcs.AIcs.CLcs.LGeess.AS
收藏

基于DNN的语音清晰度预测

Prediction of speech intelligibility with DNN-based performance measures

Carl von Ossietzky Universität Oldenburg(卡尔·冯·奥西茨基奥尔登堡大学)

Angel Mario Castro Martinez, Constantin Spille, Jana Roßbach, Birger Kollmeier, Bernd T. Meyer

摘要:

本文提出了一种基于自动语音识别(ASR)的语音可懂度模型,该模型结合了来自深度神经网络(DNN)的音素概率和从这些概率估计单词出错率的性能度量。该模型在测试过程中不需要干净的语音参考,也不需要单词标签,因为ASR解码步骤被省略,该步骤在给定音素后验概率的情况下找到最可能的单词序列。该模型通过从8个正常听力的听众的预测和观察到的语音接收阈值之间的均方根误差进行评估。识别任务包括从德国矩阵句子测试中识别噪声单词。语音材料与八个噪声掩蔽物混合,覆盖不同的调制类型,从语音形状的静态噪声到单个说话者掩蔽物。的预测性能进行比较,五个已建立的模型和一个ASR模型,使用单词标签。测试了功能和网络的两种组合。两者都包括在特征级(幅度调制滤波器组和前馈网络)或由架构(梅尔频谱图和时间延迟深度神经网络,TDNN)捕获的时间信息。TDNN模型与DNN相当,同时将参数数量减少了37倍;这种优化允许在专用助听器硬件上并行流,因为可以在每帧的10 ms内计算前向传递。该模型的性能几乎与基于标签的模型一样好,并且比基线模型产生更准确的预测。

英文摘要:

This paper presents a speech intelligibility model based on automatic speech recognition (ASR), combining phoneme probabilities from deep neural networks (DNN) and a performance measure that estimates the word error rate from these probabilities. This model does not require the clean speech reference nor the word labels during testing as the ASR decoding step, which finds the most likely sequence of words given phoneme posterior probabilities, is omitted. The model is evaluated via the root-mean-squared error between the predicted and observed speech reception thresholds from eight normal-hearing listeners. The recognition task consists of identifying noisy words from a German matrix sentence test. The speech material was mixed with eight noise maskers covering different modulation types, from speech-shaped stationary noise to a single-talker masker. The prediction performance is compared to five established models and an ASR-model using word labels. Two combinations of features and networks were tested. Both include temporal information either at the feature level (amplitude modulation filterbanks and a feed-forward network) or captured by the architecture (mel-spectrograms and a time-delay deep neural network, TDNN). The TDNN model is on par with the DNN while reducing the number of parameters by a factor of 37; this optimization allows parallel streams on dedicated hearing aid hardware as a forward-pass can be computed within the 10ms of each frame. The proposed model performs almost as well as the label-based model and produces more accurate predictions than the baseline models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑