Entity · paper

Probing Low Frame Rate Degradation in Neural Audio Codecs

paperactiveprobing-low-frame-rate-degradation-in-neural-audio-codecs-7716a0eb·1 events·first seen Jun 16, 2026

Aliases: Probing Low Frame Rate Degradation in Neural Audio Codecs

More like this (12)

quantization-induced degradation sparse frame sampling Bigger or Cheaper? Scale and Quantization Effects on Uncertainty Signals in Vision-Language Models Under Image Degradation An Evaluation Framework for Structured Audio Captions Validated by Controlled Perturbations DirectAudioEdit: Inversion-Free Text-Guided Audio Editing via Diffusion Prediction Contrast HD-Emo codec AdaCodec Leveraging Audio-LLMs to Filter Speech-to-Speech Training Data neural-ODE temporal decay AudioDER defer-to-resample REDDIT: Correcting Model-Generated Timestamp Drift in ASR without Forgetting via Replay-Based Distribution Editing

Recent events (1)

5arXiv · cs.AI·Jun 16, 2026·source ↗

Controlled ablation reveals training artifact behind low frame rate degradation in neural audio codecs

A new arXiv preprint investigates why neural audio codecs degrade sharply at low frame rates (≤6.25 Hz), a property relevant to autoregressive speech synthesis where generation cost scales with sequence length. The authors reproduce a previously reported quality cliff at 6.25 Hz and show it stems from a suboptimal training configuration—fixed clip duration starves the decoder of inter-token context at low frame rates—rather than fundamental phonemic or codebook limits. After correcting the training setup, word error rate degrades smoothly down to 1.6 Hz, suggesting low frame rate codecs are more practically accessible than prior work implied.

Inference Economics Multimodal Progress Probing Low Frame Rate Degradation in Neural Audio Codecs