语音/音频处理学术速递[11.12]

【1】HQ-SVC: Towards High-Quality Zero-Shot Singing Voice Conversion in Low-Resource Scenarios
标题：HQ-SRC：在低资源场景下实现高质量的Zero-Shot歌唱声音转换
链接：https://arxiv.org/abs/2511.08496

【2】Uncertainty Calibration of Multi-Label Bird Sound Classifiers
标题：多标签鸟声分类器的不确定度校准
链接：https://arxiv.org/abs/2511.08261

【3】Melodia: Training-Free Music Editing Guided by Attention Probing in Diffusion Models
标题：Melodia：在扩散模型中以注意力探索为指导的免训练音乐编辑
链接：https://arxiv.org/abs/2511.08252

【4】DOA Estimation with Lightweight Network on LLM-Aided Simulated Acoustic Scenes
标题：LLM辅助模拟声场景下轻量级网络的波达方向估计
链接：https://arxiv.org/abs/2511.08012

【5】Speech Emotion Recognition with Phonation Excitation Information and Articulatory Kinematics
标题：利用发音兴奋信息和关节运动学的语音情感识别
链接：https://arxiv.org/abs/2511.07955

【6】SpeechJudge: Towards Human-Level Judgment for Speech Naturalness
标题：SpeechJudge：迈向人类水平的言语自然性判断
链接：https://arxiv.org/abs/2511.07931

【7】SpikCommander: A High-performance Spiking Transformer with Multi-view Learning for Efficient Speech Command Recognition
标题：SpikCommander：一款高性能Spiking Transformer，具有多视图学习功能，用于高效的语音命令识别
链接：https://arxiv.org/abs/2511.07883

【8】SynTTS-Commands: A Public Dataset for On-Device KWS via TTS-Synthesized Multilingual Speech
标题：SynTTS-Commands：通过TTS-合成多语言语音的设备上KWS的公共数据集
链接：https://arxiv.org/abs/2511.07821

【9】Speech Separation for Hearing-Impaired Children in the Classroom
标题：课堂上听力障碍儿童的言语分离
链接：https://arxiv.org/abs/2511.07677

【10】Enabling Automatic Self-Talk Detection via Earables
标题：通过Earables启用自动自言自语检测
链接：https://arxiv.org/abs/2511.07493

【11】Quantizing Whisper-small: How design choices affect ASR performance
标题：量化Whisper-small：设计选择如何影响ASB性能
链接：https://arxiv.org/abs/2511.08093

【12】Pruning as Regularization: Sensitivity-Aware One-Shot Pruning in ASR
标题：作为规则化的修剪：ASB中敏感性意识的一次修剪
链接：https://arxiv.org/abs/2511.08092

【13】Automatic Music Mixing using a Generative Model of Effect Embeddings
标题：使用效果嵌入生成模型的自动音乐混音
链接：https://arxiv.org/abs/2511.08040

【1】Unifying Model and Layer Fusion for Speech Foundation Models
标题：语音基础模型的统一模型和层融合
链接：https://arxiv.org/abs/2511.08389

【2】Quantizing Whisper-small: How design choices affect ASR performance
标题：量化Whisper-small：设计选择如何影响ASB性能
链接：https://arxiv.org/abs/2511.08093

【3】Pruning as Regularization: Sensitivity-Aware One-Shot Pruning in ASR
标题：作为规则化的修剪：ASB中敏感性意识的一次修剪
链接：https://arxiv.org/abs/2511.08092

【4】Automatic Music Mixing using a Generative Model of Effect Embeddings
标题：使用效果嵌入生成模型的自动音乐混音
链接：https://arxiv.org/abs/2511.08040

【5】HQ-SVC: Towards High-Quality Zero-Shot Singing Voice Conversion in Low-Resource Scenarios
标题：HQ-SRC：在低资源场景下实现高质量的Zero-Shot歌唱声音转换
链接：https://arxiv.org/abs/2511.08496

【6】Melodia: Training-Free Music Editing Guided by Attention Probing in Diffusion Models
标题：Melodia：在扩散模型中以注意力探索为指导的免训练音乐编辑
链接：https://arxiv.org/abs/2511.08252