Audio Coding

Audio coding aims to efficiently represent audio signals using minimal data, balancing compression with high fidelity. Current research emphasizes neural network-based codecs, employing architectures like autoencoders, convolutional neural networks, and recurrent neural networks, often incorporating techniques such as vector quantization and generative adversarial networks to improve compression and quality at low bitrates. This field is crucial for advancing real-time communication systems, enhancing audio quality in various applications, and developing standardized benchmarks for objective and subjective evaluation of codec performance.

Papers