# F3-Tokenizer: Taming Audio Autoencoder Latents for Understanding and Generation

arXiv: https://arxiv.org/abs/2606.06357 · HTML: https://mlsanity.com/paper/2606.06357/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 21 references that are in the map.
- First posted on arXiv 4 Jun 2026.

## Builds on (most cited first, 19 of 19)

- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [SoundStream: An End-to-End Neural Audio Codec](https://mlsanity.com/paper/2107.03312/) (2021)
- [Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2301.02111/) (2023)
- [High Fidelity Neural Audio Compression](https://mlsanity.com/paper/2210.13438/) (2022)
- [AudioLM: a Language Modeling Approach to Audio Generation](https://mlsanity.com/paper/2209.03143/) (2022)
- [High-Fidelity Audio Compression with Improved RVQGAN](https://mlsanity.com/paper/2306.06546/) (2023)
- [Voicebox: Text-Guided Multilingual Universal Speech Generation at Scale](https://mlsanity.com/paper/2306.15687/) (2023)
- [Diffusion Transformers with Representation Autoencoders](https://mlsanity.com/paper/2510.11690/) (2025)
- [Codec Does Matter: Exploring the Semantic Shortcoming of Codec for Audio Language Model](https://mlsanity.com/paper/2408.17175/) (2024)
- [DiTAR: Diffusion Transformer Autoregressive Modeling for Speech Generation](https://mlsanity.com/paper/2502.03930/) (2025)
- [NextStep-1: Toward Autoregressive Image Generation with Continuous Tokens at Scale](https://mlsanity.com/paper/2508.10711/) (2025)
- [Multimodal Latent Language Modeling with Next-Token Diffusion](https://mlsanity.com/paper/2412.08635/) (2024)
- [VibeVoice Technical Report](https://mlsanity.com/paper/2508.19205/) (2025)
- [Ming-UniAudio: Speech LLM for Joint Understanding, Generation and Editing with Unified Representation](https://mlsanity.com/paper/2511.05516/) (2025)
- [SpectroStream: A Versatile Neural Codec for General Audio](https://mlsanity.com/paper/2508.05207/) (2025)
- [NEST-RQ: Next Token Prediction for Speech Self-Supervised Pre-Training](https://mlsanity.com/paper/2409.08680/) (2024)
- [SemaVoice: Semantic-Aware Continuous Autoregressive Speech Synthesis](https://mlsanity.com/paper/2605.16964/) (2026)

## Built on by (most cited first, 1 of 1)

- [FireRedAudio: A General-Purpose Audio Language Model with Decoupled Continuous Representations for Understanding and Generation](https://mlsanity.com/paper/2608.24168/) (2026)

## Abstract (arXiv)

Continuous audio autoencoders reconstruct waveforms well but often produce latents with weak structure for understanding, while self-supervised audio encoders capture semantics but are not directly decodable. This mismatch complicates a single audio tokenizer that must support both understanding and generation. We adapt continuous autoencoder latents to this setting with two components: a noise-regularized autoencoder bottleneck and a latent-side representation encoder. The bottleneck uses channel normalization and stochastic perturbation instead of KL-based variational training, yielding scale-controlled continuous latents for reconstruction and autoregressive generation. The representation encoder is trained on frozen autoencoder latents with RQ-MTP and frozen-LLM supervision. The resulting tokenizer provides high-dimensional representations for understanding while preserving normalized continuous latents as generation targets