# FlexiCodec: A Dynamic Neural Audio Codec for Low Frame Rates

arXiv: https://arxiv.org/abs/2510.00981 · HTML: https://mlsanity.com/paper/2510.00981/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Oct 2025.

## Built on by (most cited first, 10 of 10)

- [Beyond Fixed Frames: Dynamic Character-Aligned Speech Tokenization](https://mlsanity.com/paper/2601.23174/) (2026)
- [Reconstruct! Don't Encode: Self-Supervised Representation Reconstruction Loss for High-Intelligibility and Low-Latency Streaming Neural Audio Codec](https://mlsanity.com/paper/2603.05887/) (2026)
- [DeepASMR: LLM-Based Zero-Shot ASMR Speech Generation for Anyone of Any Voice](https://mlsanity.com/paper/2601.15596/) (2026)
- [Probing Low Frame Rate Degradation in Neural Audio Codecs](https://mlsanity.com/paper/2606.16969/) (2026)
- [Elastic Time: Dynamic Frame Rate Bottlenecks for Neural Audio Coding](https://mlsanity.com/paper/2606.27320/) (2026)
- [ZipCodec: Ultra-Low-Frame-Rate Streaming Speech Coding](https://mlsanity.com/paper/2609.11642/) (2026)
- [TokenMapper: A Step Toward Interoperable Speech Token Translation](https://mlsanity.com/paper/2609.12563/) (2026)
- [LACE: Layer-Wise Compression for Dynamic Frame Rate Codecs](https://mlsanity.com/paper/2609.17509/) (2026)
- [Q-SPT: Learnable Query-Based Compression for Low-Frame-Rate Speech Tokenization](https://mlsanity.com/paper/2610.01492/) (2026)
- [GS-Codec: A Gaussian-Splatting Bottleneck for Neural Audio Coding](https://mlsanity.com/paper/2610.04651/) (2026)

## Abstract (arXiv)

Neural audio codecs are foundational to speech language models. It is expected to have a low frame rate and decoupled semantic and acoustic information. A lower frame rate codec can reduce the computational cost of speech language models by shortening the sequence length. Recent studies have developed 12.5Hz low-frame-rate audio codecs, but even lower frame rate codecs remain underexplored. We find that a major challenge for very low frame rate tokens is missing semantic information. This paper introduces FlexiCodec to address this limitation. FlexiCodec improves semantic preservation with a dynamic frame rate approach and introduces a novel architecture featuring an ASR feature-assisted dual stream encoding and Transformer bottlenecks. With dynamic frame rates, it uses less frames at information-sparse regions through adaptively merging semantically similar frames. A dynamic frame rate also allows FlexiCodec to support inference-time controllable frame rates between 3Hz and 12.5Hz. Experiments on 6.25Hz, 8.3Hz and 12.5Hz average frame rates confirm that FlexiCodec excels over baseline systems in semantic information preservation and delivers a high audio reconstruction quality. We also validate the effectiveness of FlexiCodec in language model-based TTS. Demos are available at: https://flexicodec.github.io. Code is available at: https://github.com/amphionteam/flexicodec.