arXiv每日推荐-5.9:语音/音频每日论文速递

同步公众号(arXiv每日学术速递)

【1】 The Perceptimatic English Benchmark for Speech Perception Models

标题：言语感知模型的感性英语基准

作者： Juliette Millet, Ewan Dunbar

备注：Accepted to CogSci Conference 2020

链接：https://arxiv.org/abs/2005.03418

【2】 Crop Aggregating for short utterances speaker verification using raw waveforms

标题：使用原始波形的短话语的裁剪聚集说话人验证

作者： Seung-bin Kim, Ha-Jin Yu

链接：https://arxiv.org/abs/2005.03329

【3】 Cotatron: Transcription-Guided Speech Encoder for Any-to-Many Voice Conversion without Parallel Data

标题：Coatron：转录导向语音编码器，用于无并行数据的任意对多语音转换

作者： Seung-won Park, Myun-chul Joe

备注：Submitted to Interspeech 2020

链接：https://arxiv.org/abs/2005.03295

【4】 ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Context

标题：ContextNet：用于全局上下文自动语音识别的改进卷积神经网络

作者： Wei Han, Yonghui Wu

链接：https://arxiv.org/abs/2005.0319

【5】 Study of human phonation in a full body domain

标题：人体全身发声的研究

作者： Shakti Saurabh, Daniel Bodony

链接：https://arxiv.org/abs/2005.02168

【6】 End-to-end Whispered Speech Recognition with Frequency-weighted Approaches and Layer-wise Transfer Learning

标题：基于频率加权和分层转移学习的端到端耳语语音识别

作者： Heng-Jui Chang, Lin-shan Lee

备注：submitted to INTERSPEECH 2020

链接：https://arxiv.org/abs/2005.0197

【7】 Study of human phonation in a full body domain

标题：人体全身发声的研究

作者： Shakti Saurabh, Daniel Bodony

链接：https://arxiv.org/abs/2005.02168

【8】 End-to-end Whispered Speech Recognition with Frequency-weighted Approaches and Layer-wise Transfer Learning

标题：基于频率加权和分层转移学习的端到端耳语语音识别

作者： Heng-Jui Chang, Lin-shan Lee

备注：submitted to INTERSPEECH 2020

链接：https://arxiv.org/abs/2005.0197