CORTEXA
← Browse
arxivcs.CLcs.AI2026-07-09

When Synthetic Speech Is All You Have: Better Call GRPO

Shashi Kumar, Yanis Labrak, Hasindri Watawana, Sergio Burdisso, Esaú Villatoro-Tello, Kadri Hacioğlu, Petr Motlicek, Andreas Stolcke

LLM-based ASR adapted to regulated domains such as banking is bottlenecked by privacy: real speech is costly and legally constrained to collect, making synthetic text-to-speech (TTS) an attractive substitute. Yet synthetic speech stays acoustically mismatched with real recordings, and work on this gap has stayed within supervised fine-tuning (SFT). We instead turn to reinforcement learning, and show that Group Relative Policy Optimization (GRPO) extracts far more from the same synthetic speech than SFT. Synthetic-only adaptation of the model with GRPO, a critic-free method rewarding low-WER hypotheses, reduces WER by 40\% relative to SFT (36.71\%$\to$22.09\%), and an SFT-then-GRPO combination pushes this further to 45\%. We trace the gain to behavior rather than representation: GRPO reduces insertion errors by improving stopping calibration and speech-to-text alignment by better anchoring attention to audio, leaving early-layer representations intact. When synthetic speech is the main resource, reinforcement learning should be preferred over supervised fine-tuning.

View free PDFSource page

Related papers

arxivcs.CLcs.AI2026-06-27

How to Leverage Synthetic Speech for LLM-Based ASR Systems?

Yanis Labrak, Dairazalia Sanchez-Cortes, Sergio Burdisso, Séverin Baroudi, Shashi Kumar, Esaú Villatoro-Tello, et al.

In regulated domains such as banking and healthcare, where privacy constraints make real speech costly to collect and retain, synthetic speech from modern text-to-speech (TTS) is an appealing alternative for training automatic speech recognition (ASR) without exposing sensitive c…

View free PDFSource page
arxivcs.CLcs.AIeess.AS2026-07-06

SPEARBench: A Benchmark for Naturalness Evaluation in Streaming Speech-to-Speech Language Models

Thomas Thebaud, Yuzhe Wang, Hao Zhang, Sathvik Manikantan Napa Ugandhar, Ashish Hallur, Georgi Tinchev, et al.

Streaming speech-to-speech language models aim to answer spoken queries directly with synthetic speech. However, standard speech and text benchmarks do not capture whether these systems behave naturally in conversations, where timing, turn-taking, prosody, interpersonal stance, l…

View free PDFSource page
arxivcs.AIcs.CLcs.CVcs.MM2026-07-14

Do We Really Need Multimodal Emotion Language Models Larger Than 1B Parameters?

Kaiwen Zheng, Junchen Fu, Wenhao Deng, Hu Han, Joemon M. Jose, Xuri Ge

Recent advances in multimodal large language models (MLLMs) have significantly improved the performance of multimodal emotion recognition (MER) and enabled interpretable description generation by jointly modeling video, audio, and language, etc. However, these performance improve…

View free PDFSource page
arxivcs.CLcs.AI2026-06-29

TRACE: Temporal Relationship-Aware Conversational Entrainment Detection in Dyadic Speech

Sathvik Manikantan Napa Ugandhar, Hao Zhang, Alison Gunzler, Yuzhe Wang, Thomas Thebaud, Georgi Tinchev, et al.

With the proliferation of speech AI agents, understanding emotional entrainment in conversational interaction has become increasingly important. Emotional entrainment is shaped by social relationships and conversational context, influencing affective coordination over time. We in…

View free PDFSource page
arxivcs.SDcs.AIcs.CLeess.AS2026-06-26

DG^VoiC: Speaker Clustering for Fraud Investigation under Real Call-Centre Conditions

Muhammad Shakeel Akram, Amal Htait, Abdul Hamid Sadka, Emma Meisingseth, Karishma Jaitly

Insurance fraud remains costly and operationally difficult, particularly in call-centre workflows where many customer interactions begin at FNOL. While recent fraud detection methods mainly rely on structured data, text, or images, repeated speaker identity across calls remains u…

View free PDFSource page