# RLAIF-SPA: Structured AI Feedback for Semantic-Prosodic Alignment in Speech Synthesis

arXiv: https://arxiv.org/abs/2510.14628 · HTML: https://mlsanity.com/paper/2510.14628/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Preference alignment for recommenders and RAG, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 25 of its 34 references that are in the map.
- First posted on arXiv 16 Oct 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Preference alignment for recommenders and RAG](https://mlsanity.com/alignment/preference-alignment-for-recommenders-and-rag/)

## Builds on (most cited first, 25 of 25)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversations](https://mlsanity.com/paper/1810.02508/) (2018)
- [emotion2vec: Self-Supervised Pre-Training for Speech Emotion Representation](https://mlsanity.com/paper/2312.15185/) (2023)
- [Emotional Voice Conversion: Theory, Databases and ESD](https://mlsanity.com/paper/2105.14762/) (2021)
- [SpeechAlign: Aligning Speech Generation to Human Preferences](https://mlsanity.com/paper/2404.05600/) (2024)
- [EmoBox: Multilingual Multi-corpus Speech Emotion Recognition Toolkit and Benchmark](https://mlsanity.com/paper/2406.07162/) (2024)
- [EmoDiff: Intensity Controllable Emotional Text-to-Speech with Soft-Label Guidance](https://mlsanity.com/paper/2211.09496/) (2022)
- [MusicRL: Aligning Music Generation to Human Preferences](https://mlsanity.com/paper/2402.04229/) (2024)
- [Emo-DPO: Controllable Emotional Speech Synthesis through Direct Preference Optimization](https://mlsanity.com/paper/2409.10157/) (2024)
- [Emotional speech synthesis with rich and granularized control](https://mlsanity.com/paper/1911.01635/) (2019)
- [EmoSphere-TTS: Emotional Style and Intensity Modeling via Spherical Emotion Vector for Controllable Emotional Text-to-Speech](https://mlsanity.com/paper/2406.07803/) (2024)
- [Align-SLM: Textless Spoken Language Models with Reinforcement Learning from AI Feedback](https://mlsanity.com/paper/2411.01834/) (2024)
- [A Vector Quantized Approach for Text to Speech Synthesis on Real-World Spontaneous Speech](https://mlsanity.com/paper/2302.04215/) (2023)
- [EmoMix: Emotion Mixing via Diffusion Models for Emotional Speech Synthesis](https://mlsanity.com/paper/2306.00648/) (2023)
- [Laugh Now Cry Later: Controlling Time-Varying Emotional States of Flow-Matching-Based Zero-Shot Text-to-Speech](https://mlsanity.com/paper/2407.12229/) (2024)
- [ControlSpeech: Towards Simultaneous and Independent Zero-shot Speaker Cloning and Zero-shot Language Style Control](https://mlsanity.com/paper/2406.01205/) (2024)
- [EmoSpeech: Guiding FastSpeech2 Towards Emotional Text to Speech](https://mlsanity.com/paper/2307.00024/) (2023)
- [ZET-Speech: Zero-shot adaptive Emotion-controllable Text-to-Speech Synthesis with Diffusion and Style-based Models](https://mlsanity.com/paper/2305.13831/) (2023)
- [MM-TTS: Multi-modal Prompt based Style Transfer for Expressive Text-to-Speech Synthesis](https://mlsanity.com/paper/2312.10687/) (2023)
- [SageLM: A Multi-aspect and Explainable Large Language Model for Speech Judgement](https://mlsanity.com/paper/2508.20916/) (2025)
- [Group Relative Policy Optimization for Text-to-Speech with Large Language Models](https://mlsanity.com/paper/2509.18798/) (2025)
- [AV-EmoDialog: Chat with Audio-Visual Users Leveraging Emotional Cues](https://mlsanity.com/paper/2412.17292/) (2024)

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

- [Do Audio LLMs Listen or Read? Analyzing and Mitigating Paralinguistic Failures with VoxParadox](https://mlsanity.com/paper/2605.27772/) (2026)

## Abstract (arXiv)

Recent advances in Text-To-Speech (TTS) synthesis have achieved near-human speech quality in neutral speaking styles. However, most existing approaches either depend on costly emotion annotations or optimize surrogate objectives that fail to adequately capture perceptual emotional quality. As a result, the generated speech, while semantically accurate, often lacks expressive and emotionally rich characteristics. To address these limitations, we propose RLAIF-SPA, a novel framework that integrates Reinforcement Learning from AI Feedback (RLAIF) to directly optimize both emotional expressiveness and intelligibility without human supervision. Specifically, RLAIF-SPA incorporates Automatic Speech Recognition (ASR) to provide semantic accuracy feedback, while leveraging structured reward modeling to evaluate prosodic-emotional consistency. RLAIF-SPA enables more precise and nuanced control over expressive speech generation along four structured evaluation dimensions: Structure, Emotion, Speed, and Tone. Extensive experiments on Libri-Speech, MELD, and Mandarin ESD datasets demonstrate consistent gains across clean read speech, conversational dialogue, and emotional speech. On Libri-Speech, RLAIF-SPA consistently outperforms Chat-TTS, achieving a 26.1% reduction in word error rate, a 9.1% improvement in SIM-O, and over 10% gains in human subjective evaluations.