[
    {
        "id": "osp-23301",
        "type": "article-journal",
        "title": "Spoken Language Models that Think Aloud",
        "author": [
            {
                "family": "Ao",
                "given": "Junyi"
            },
            {
                "family": "Peng",
                "given": "Kainan"
            },
            {
                "family": "Ma",
                "given": "Mingbo"
            },
            {
                "family": "Zhang",
                "given": "Shun"
            },
            {
                "family": "Tang",
                "given": "Zhenyu"
            },
            {
                "family": "Ma",
                "given": "Xutai"
            },
            {
                "family": "Li",
                "given": "Xiang"
            },
            {
                "family": "Li",
                "given": "Yinghao"
            },
            {
                "family": "Wang",
                "given": "Yuancheng"
            },
            {
                "family": "Wu",
                "given": "Zhizheng"
            },
            {
                "family": "Li",
                "given": "Haizhou"
            },
            {
                "family": "He",
                "given": "Qing"
            },
            {
                "family": "Liu",
                "given": "Xubo"
            }
        ],
        "URL": "https://omanscience.com/en/articles/spoken-language-models-that-think-aloud",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "While Chain-of-Thought (CoT) reasoning has improved the capability of language models, directly applying it to Spoken Language Models (SLMs) may introduce long silent intervals under the serial \"think-then-speak\" paradigm, disrupting real-time spoken interaction. To address this issue, we propose an asynchronous think-aloud framework for reasoning-based SLMs within the Thinker-Talker architecture. The framework maintains a primary reasoning stream for logical deduction and a lightweight think-aloud stream that generates short, task-grounded progress utterances conditioned on the user input and the evolving reasoning state. A dynamic balance strategy coordinates the two streams at runtime, triggering additional think-aloud speech to avoid silent gaps and canceling pending utterances when the final response becomes ready. Experiments on spoken reasoning and question-answering benchmarks show that our approach substantially reduces user-audible silence during reasoning while maintaining answer accuracy comparable to that of a serial \"think-then-speak\" baseline, demonstrating the potential of asynchronous think-aloud for responsive interaction in SLMs."
    }
]