RubatoCodec: variable-frame-rate neural audio codec open-sourced with weights, code and paper
realmrfakename · x · 2026-10-28
Developer AratakoLM has released RubatoCodec, a variable-frame-rate neural audio codec, fully open-sourced with weights, training and inference code, and an accompanying paper.
Designed primarily for practical TTS use, the codec reportedly achieves notably strong performance on English and Japanese, serving as an acoustic intermediate representation in speech synthesis pipelines.
More from Research
- REGER paper adds explicit item relations to generative recommendation via semantic IDs — _reachsumit · 2026-10-29
- ARC Prize finds v2 = v1^4.53 power curve explains 98% of ARC-AGI-2 score variance — GregKamradt · 2026-10-29
- DiscoTrace paper on human vs LLM answering strategies wins Best Paper at EMNLP 2026 — rachelrudinger · 2026-10-29
- A Red Heart Image Makes Claude and GPT Write Recursive Code 100% of the Time, DeepMind Finds — dair_ai · 2026-10-29
- CMASS: New agentic search benchmark with 423 QA pairs over 550M docs — lintool · 2026-10-29
- Proteins as strings: Profluent uses language models to design gene editors — nathanbenaich · 2026-10-29