Audio processing method, apparatus, device, and storage medium
By using multi-channel microphones to collect audio data and training audio models with differentiated attenuation processing, the problem of insufficient naturalness and realism in audio processing in complex acoustic environments is solved, achieving the effect of accurately extracting target audio and preserving spatial information in complex environments.
CN122417048APending Publication Date: 2026-07-17GUANGZHOU SHIYUAN ELECTRONICS CO LTD +1
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- CN202510076165.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-17
- Publication Date
- 2026-07-17
Smart Images
Figure CN122417048A_ABST
Abstract
本申请涉及一种音频处理方法、装置、设备及存储介质，所述方法获取拾音目标和不同方位的噪音目标的多通道音频，基于噪音目标与拾音目标之间的方位信息，确定每个噪音目标的第一衰减系数，根据该衰减系数对相应的多通道噪音音频进行衰减处理。将原始的多通道噪音音频与多通道目标音频进行多通道音频合成，得到多通道合成音频，作为训练样本。同时，将衰减处理后的多通道噪音音频与多通道目标音频进行音频合成得到单通道合成音频，作为训练样本对应的标签数据。最终，将训练样本和对应的标签数据输入预构建的音频处理模型进行训练，训练得到的音频处理模型。通过训练得到的音频处理模型处理多通道会议音频，能够得到更加自然和真实的目标音频。
Need to check novelty before this filing date? Find Prior Art