PEER-REVIEWED PAPERNeural Audio CodecsDOI: 10.1109/TASLP.2021.3129994
SoundStream: An End-to-End Neural Audio Codec
IEEE/ACM Transactions on Audio, Speech, and Language Processing2021-12-08
Research Abstract & Summary
Peer-reviewed paper introducing SoundStream, an end-to-end neural audio codec utilizing convolutional encoders, residual vector quantization (RVQ), and adversarial discriminators to efficiently discretize acoustic signals into multi-rate token sequences.
Scholarly Resolver: Canonical literature access