# Unified Audio Intelligence Without Regressing on Text Intelligence

arXiv: https://arxiv.org/abs/2607.05196 · HTML: https://mlsanity.com/paper/2607.05196/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 36 references that are in the map.
- First posted on arXiv 6 Jul 2026.

## Builds on (most cited first, 5 of 5)

- [Recent Advances in Bayesian Optimization](https://mlsanity.com/paper/2206.03301/) (2022)
- [ADO-LLM: Analog Design Bayesian Optimization with In-Context Learning of Large Language Models](https://mlsanity.com/paper/2406.18770/) (2024)
- [An Efficient Batch Constrained Bayesian Optimization Approach for Analog Circuit Synthesis via Multi-objective Acquisition Ensemble](https://mlsanity.com/paper/2106.15412/) (2021)
- [Area Optimisation of Two Stage Miller Compensated Op-Amp in 65 nm Using Hybrid PSO](https://mlsanity.com/paper/2106.09383/) (2021)
- [Constrained Bayesian Optimization Using a Lagrange Multiplier Applied to Power Transistor Design](https://mlsanity.com/paper/2308.09612/) (2023)

## 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)

Audio intelligence involves understanding, reasoning about, and generating both audio and speech. In this work, we introduce Nemotron-Labs-Audex-30B-A3B (Audex), a unified audio-text LLM built on Nemotron-Cascade-2-30B-A3B, a strong text-only MoE LLM. Audex adopts a simple unified design with a single Transformer decoder: audio inputs are encoded and projected into the text embedding space, while text tokens and quantized audio output tokens are treated uniformly during generation. This architecture enables strong audio-text fusion, seamless multimodal generation, and compatibility with standard LLM training and inference infrastructure. For training, we meticulously curate audio-text datasets comprising 157.4B audio tokens and 320.5B text tokens. We apply multi-stage supervised training on these datasets, followed by text-only Cascade RL and multi-domain on-policy distillation. Audex delivers state-of-the-art audio understanding, speech recognition and translation, text-to-speech, audio generation, and speech-to-speech generation, while preserving very compelling reasoning, alignment, knowledge, long-context, and agentic capabilities of its text-only LLM backbone with marginal or no regression. We release the model checkpoints to facilitate open research.