paper-with-me

홈 › Papers

INSPIRE: A Benchmark for Instruction-Aware Speech Retrieval

2026-08-17 · Chen-An Li, Hung-yi Lee arxiv

Existing speech retrieval systems rely on fixed similarity matching and cannot adapt to diverse user intents. We introduce INSPIRE, the first benchmark for instruction-aware speech retrieval, in which natural-language instructions dynamically specify relevance criteria, including semantic content, speaker identity, speaking style, environmental sounds, and their combinations. We evaluate four retrieval paradigms: large audio-language models, cascaded pipelines, self-supervised speech models, and contrastive audio-language models. Our results reveal that no current method robustly handles all retrieval intents. Text-based approaches perform relatively better at semantic retrieval but struggle with paralinguistic attributes, while speech-based models are moderately better at capturing acoustic properties but falter at following instructions. These findings highlight the need for unified architectures capable of instruction-aware speech retrieval.

📄 PDF Abstract BibTeX arXiv:2608.16203

Code (0)

등록된 구현이 없습니다.

Tasks

Semantic Retrieval

Similar Papers 제목 키워드 기반

ALM2Vec: Learning Audio Embeddings for Universal Audio Retrieval with Large Audio-Language Models

2026-06-27 · Fengjie Lu, Chenang Jiang, Jiarui Hai, Helin Wang 외 arxiv

Recent advances in language--audio retrieval have been largely driven by contrastive dual-encoder architectures that align audio and text in a shared embedding space. While effective, existing retrieval embeddings are pr…

Question Answering

Speech-IFEval: Evaluating Instruction-Following and Quantifying Catastrophic Forgetting in Speech-Aware Language Models

2025-05-25 · Ke-Han Lu, Chun-Yi Kuan, Hung-Yi Lee

We introduce Speech-IFeval, an evaluation framework designed to assess instruction-following capabilities and quantify catastrophic forgetting in speech-aware language models (SLMs). Recent SLMs integrate speech percepti…

Instruction Following

Task-aware Retrieval with Instructions

2022-11-16 · Akari Asai, Timo Schick, Patrick Lewis, Xilun Chen 외

We study the problem of retrieval with instructions, where users of a retrieval system explicitly describe their intent along with their queries. We aim to develop a general-purpose task-aware retrieval system using mult…

Retrieval

AgentSteerTTS: A Multi-Agent Closed-Loop Framework for Composite-Instruction Text-to-Speech

2026-05-14 · Bin Kang, Shaoguo Wen, Yang Fan, Shunlong Wu 외 arxiv

While existing text-to-speech (TTS) models exhibit high expressiveness, fine-grained control over composite instructions remains challenging due to the structural mismatch between discrete textual intents and continuous …

Continuous Control

Beyond Content Relevance: Evaluating Instruction Following in Retrieval Models

2024-10-31 · Jianqun Zhou, Yuanlei Zheng, Wei Chen, Qianqian Zheng 외

Instruction-following capabilities in LLMs have progressed significantly, enabling more complex user interactions through detailed prompts. However, retrieval systems have not matched these advances, most of them still r…

Instruction FollowingRerankingRetrieval