3 papers
cs.SD2026
wav2tok 2.0: Scalable Audio Tokenization Maintaining Explicit Pairwise Token Alignment for Efficient Audio Retrieval
Adhiraj Banerjee, Vipul Arora
Learning discrete speech representations that preserve similarity across variable-length utterances is central to query-by-example spoken term detection (QbE-STD). While wav2tok in…
cs.LG2026
PairAlign: A Framework for Sequence Tokenization via Self-Alignment with Applications to Audio Tokenization
Adhiraj Banerjee, Vipul Arora
Modern learning systems represent perceptual signals with continuous vectors, but comparison, retrieval, memory, alignment, and reasoning are often naturally symbolic. In language,…
cs.SD2026
CodecSep: Prompt-Driven Universal Sound Separation on Neural Audio Codec Latents
Adhiraj Banerjee, Vipul Arora
Text-guided sound separation enables flexible audio editing, assistive listening, and open-domain source extraction, but systems such as AudioSep remain too expensive for low-laten…