AI & ML
Robustness of Text-Only Adapted Speech LLMs for Spoken Semantic Parsing
arXiv cs.HC0 views
arXiv:2606.21453v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used to generate structured outputs from natural language instructions, including semantic frames and function calls. When the input is speech, this task becomes spoken semantic parsing, which can be performed using either Speech LLMs or ASR-LLM cascades. Although both approaches can be adapted with text-only task supervision, their relative robustness to acoustic degradation remains unclear. We compare these approaches under matched text-form supervision using 3B and 7B models across three datasets, evaluated on clean speech and babble noise at five SNR levels. Results show that Speech LLMs generally achieve better performance in command understanding, intent prediction, and function calling, particularly under severe noise, while improvements in fine-grained entity prediction are less consistent. An additional evaluation shows that Speech LLMs achieve lower average inference time per sample than ASR-LLM cascades at both model scales.
Source:https://arxiv.org/abs/2606.21453PreviousLLM-Assisted Reranking to Operationalize Nuanced Objectives in Recommender SystemsNextFrom Tasks to Turns: Measuring Human Labor Contributions in Human-Agent Joint Production
Related posts
Our health check said "db ok" for 40 minutes. It was talking to a cache.
0dev.to WebDev
Banco do Brasil: acesso indevido pode ter exposto CPF, e-mail e dados bancários
0dev.to Security

When the Attacker Has No Return Address
0dev.to Security
La nube te mantiene responsable de todos modos; lo local simplemente nunca comet
0dev.to Security
Interrupções em bancos e redes da NSA são ligadas a correções de falhas por IA
0dev.to Security
Reverse Path Filtering Is Not a Substitute for Routing Policy
0dev.to Linux