Elastic Time: Dynamic Frame Rate Bottlenecks for Neural Audio Coding
For audio compression and generation tasks, this method offers a flexible way to adjust temporal resolution, potentially improving efficiency in downstream modeling.
Elastic Time introduces a dynamic frame-rate bottleneck for neural audio autoencoders, enabling variable temporal resolution to improve efficiency-quality tradeoffs. Experiments show deployment-time rate control with better tradeoffs than fixed-frame-rate baselines.
Neural audio autoencoders have become a core component of compression, feature extraction, and generation. However, while existing systems support variable bitrate, the vast majority of models still operate at a fixed latent frame-rate, allocating equal temporal budget to regions with very different information density, which can result in unnecessarily long sequences. We introduce Elastic Time, a dynamic frame-rate bottleneck that converts fixed-frame-rate autoencoders to dynamic ones. Our method learns a lightweight latent predictor used to decide which frames can be skipped and later reconstructed, enabling efficient greedy boundary selection at inference. Experiments show our method enables deployment-time rate control while improving efficiency-quality tradeoffs relative to baselines. Overall, we provide a flexible mechanism for adjusting temporal resolution in audio autoencoders, potentially facilitating more efficient downstream modeling for generation and long-context tasks.