When Synthetic Speech Is All You Have: Better Call GRPO
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 t
Record details
Published: 9 July 2026
Source: arXiv
Category: Research
Topics: Regulation · Privacy · Finance, VC & PE
Retrieved: 14 July 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
Explainable Graph Neural Networks for Interbank Contagion Surveillance: A Regulatory-Aligned Framework for the U.S. Banking Sector
arXiv · 14 April 2026
I investigated Palantir’s foothold in the British state – and what I found should worry us all | Peter Geoghegan
The Guardian · 15 July 2026
Morning Docket: 07.20.26
Above the Law (legal tech) · 20 July 2026
TikTok is violating EU regulations on kids’ privacy rights, according to an investigation
Fast Company Tech · 24 July 2026
LLM-assisted formalization for deterministic detection of statutory inconsistency in tax law
Artificial Intelligence and Law · 10 July 2026
Privacy Detective: A Narrative Game that Cultivates Student Developers' Privacy Awareness by Harnessing Legal Documents
arXiv · 10 July 2026
How to cite this record
ethics.ai (9 July 2026), “When Synthetic Speech Is All You Have: Better Call GRPO,” evidence record 90, https://ethics.ai/record/90 (originally published by arXiv).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.