cd /news/artificial-intelligence/bridging-the-stability-expressivity-… · home topics artificial-intelligence article
[ARTICLE · art-16059] src=arxiv.org pub= topic=artificial-intelligence verified=true sentiment=↑ positive

Bridging the Stability-Expressivity Gap: Synthetic Data Scaling and Preference Alignment for Low-Resource Spoken Language Models

Researchers at the University of Cambridge have identified a fundamental trade-off between stability and expressivity in spoken language models trained on synthetic data, where phonetic accuracy improves but prosodic variability collapses. To address this, they developed two self-alignment frameworks—Disentanglement-Guided Self-Alignment and Temperature-Driven Self-Critique—that outperform commercial systems like ElevenLabs and Gemini Pro. The breakthrough enables the first zero-shot voice cloning capability for Lao, a low-resource language.

read1 min publishedMay 28, 2026

arXiv:2605.27383v1 Announce Type: new Abstract: Spoken Language Models (SLMs) have emerged as a promising paradigm for speech synthesis by bypassing explicit grapheme-to-phoneme pipelines. However, their effectiveness in low-resource languages remains fundamentally limited by the scarcity of transcribed speech. In practice, synthetic data has become the primary strategy for scaling SLMs in such settings, providing reliable phonetic supervision when real data is insufficient. In this work, we show that this reliance introduces a fundamental trade-off, which we term the Stability-Expressivity Gap: while synthetic data improves phonetic accuracy, it progressively suppresses prosodic variability, ultimately leading to a collapse of expressivity (Synthetic Erosion). To bridge this gap, we propose two self-alignment frameworks. Disentanglement-Guided Self-Alignment (DGSA) recovers expressivity for complex languages by exploiting prosody-timbre separation. For regimes where authentic references are exceptionally limited, Temperature-Driven Self-Critique (TDSC) stabilizes generation through automated exploration and filtering. Our approach outperforms strong commercial systems, including ElevenLabs and Gemini Pro, and enables the first zero-shot voice cloning capability for Lao.

── more in #artificial-intelligence 4 stories · sorted by recency
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/bridging-the-stabili…] indexed:0 read:1min 2026-05-28 ·