JEPA as a Neural Tokenizer: Learning Robust Speech Representations with Density Adaptive Attention
By: Georgios Ioannides , Christos Constantinou , Aman Chadha and more
Potential Business Impact:
Makes computers understand speech better, faster.
We introduce a two-stage self-supervised framework that combines the Joint-Embedding Predictive Architecture (JEPA) with a Density Adaptive Attention Mechanism (DAAM) for learning robust speech representations. Stage~1 uses JEPA with DAAM to learn semantic audio features via masked prediction in latent space, fully decoupled from waveform reconstruction. Stage~2 leverages these representations for efficient tokenization using Finite Scalar Quantization (FSQ) and a mixed-radix packing scheme, followed by high-fidelity waveform reconstruction with a HiFi-GAN decoder. By integrating Gaussian mixture-based density-adaptive gating into the JEPA encoder, the model performs adaptive temporal feature selection and discovers hierarchical speech structure at a low frame rate of 2.5~Hz. The resulting tokens (47.5 tokens/sec) provide a reversible, highly compressed, and language-model-friendly representation that is competitive with, and often more efficient than, existing neural audio codecs.
Similar Papers
Gaussian Embeddings: How JEPAs Secretly Learn Your Data Density
Machine Learning (CS)
Helps computers understand how likely things are.
SparseJEPA: Sparse Representation Learning of Joint Embedding Predictive Architectures
Machine Learning (CS)
Makes AI understand pictures better and more clearly.
WavJEPA: Semantic learning unlocks robust audio foundation models for raw waveforms
Sound
Makes computers understand sounds faster and better.