# Investigating Group Relative Policy Optimization for Diffusion Transformer based Text-to-Audio Generation

arXiv: https://arxiv.org/abs/2603.01565 · HTML: https://mlsanity.com/paper/2603.01565/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Audio, speech and music, one of 14 directions in Diffusion models.
- Builds on 52 of its 55 references that are in the map.
- First posted on arXiv 2 Mar 2026.

## Placed in

- Diffusion models / [Audio, speech and music](https://mlsanity.com/diffusion/audio-speech-and-music/)

## Builds on (most cited first, 30 of 52)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](https://mlsanity.com/paper/2405.04434/) (2024)
- [Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2301.02111/) (2023)
- [Flow-GRPO: Training Flow Matching Models via Online RL](https://mlsanity.com/paper/2505.05470/) (2025)
- [Qwen3-Omni Technical Report](https://mlsanity.com/paper/2509.17765/) (2025)
- [AudioLDM 2: Learning Holistic Audio Generation with Self-supervised Pretraining](https://mlsanity.com/paper/2308.05734/) (2023)
- [Diffsound: Discrete Diffusion Model for Text-to-sound Generation](https://mlsanity.com/paper/2207.09983/) (2022)
- [NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers](https://mlsanity.com/paper/2304.09116/) (2023)
- [NaturalSpeech 3: Zero-Shot Speech Synthesis with Factorized Codec and Diffusion Models](https://mlsanity.com/paper/2403.03100/) (2024)
- [Stable Audio Open](https://mlsanity.com/paper/2407.14358/) (2024)
- [Speak Foreign Languages with Your Own Voice: Cross-Lingual Neural Codec Language Modeling](https://mlsanity.com/paper/2303.03926/) (2023)
- [NaturalSpeech: End-to-End Text to Speech Synthesis with Human-Level Quality](https://mlsanity.com/paper/2205.04421/) (2022)
- [E2 TTS: Embarrassingly Easy Fully Non-Autoregressive Zero-Shot TTS](https://mlsanity.com/paper/2406.18009/) (2024)
- [VALL-E 2: Neural Codec Language Models are Human Parity Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2406.05370/) (2024)
- [Qwen3-TTS Technical Report](https://mlsanity.com/paper/2601.15621/) (2026)
- [Step-Audio 2 Technical Report](https://mlsanity.com/paper/2507.16632/) (2025)
- [Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization](https://mlsanity.com/paper/2404.09956/) (2024)
- [TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching and Clap-Ranked Preference Optimization](https://mlsanity.com/paper/2412.21037/) (2024)
- [Autoregressive Speech Synthesis without Vector Quantization](https://mlsanity.com/paper/2407.08551/) (2024)
- [Read, Watch and Scream! Sound Generation from Text and Video](https://mlsanity.com/paper/2407.05551/) (2024)
- [VALL-E R: Robust and Efficient Zero-Shot Text-to-Speech Synthesis via Monotonic Alignment](https://mlsanity.com/paper/2406.07855/) (2024)
- [Text-to-Audio Generation Synchronized with Videos](https://mlsanity.com/paper/2403.07938/) (2024)
- [LAFMA: A Latent Flow Matching Model for Text-to-Audio Generation](https://mlsanity.com/paper/2406.08203/) (2024)
- [F5R-TTS: Improving Flow-Matching based Text-to-Speech with Group Relative Policy Optimization](https://mlsanity.com/paper/2504.02407/) (2025)
- [AudioLCM: Text-to-Audio Generation with Latent Consistency Models](https://mlsanity.com/paper/2406.00356/) (2024)
- [FreeAudio: Training-Free Timing Planning for Controllable Long-Form Text-to-Audio Generation](https://mlsanity.com/paper/2507.08557/) (2025)
- [FlashAudio: Rectified Flows for Fast and High-Fidelity Text-to-Audio Generation](https://mlsanity.com/paper/2410.12266/) (2024)

## Abstract (arXiv)

Text-to-audio (T2A) generation has advanced considerably in recent years, yet existing methods continue to face challenges in accurately rendering complex text prompts, particularly those involving intricate audio effects, and achieving precise text-audio alignment. While prior approaches have explored data augmentation, explicit timing conditioning, and reinforcement learning, overall synthesis quality remains constrained. In this work, we experiment with reinforcement learning to further enhance T2A generation quality, building on diffusion transformer (DiT)-based architectures. Our method first employs a large language model (LLM) to generate high-fidelity, richly detailed audio captions, substantially improving text-audio semantic alignment, especially for ambiguous or underspecified prompts. We then apply Group Relative Policy Optimization (GRPO), a recently introduced reinforcement learning algorithm, to fine-tune the T2A model. Through systematic experimentation with diverse reward functions (including CLAP, KL, FAD, and their combinations), we identify the key drivers of effective RL in audio synthesis and analyze how reward design impacts final audio quality. Experimental results demonstrate that GRPO-based fine-tuning yield substantial gains in synthesis fidelity and prompt adherence.