Human-1 by Josh Talks: 基于真实对话的印地语全双工对话建模框架
Human-1 by Josh Talks: A Full-Duplex Conversational Modeling Framework in Hindi using Real-World Conversations
AI总结：
本文通过适配Moshi架构，使用自定义印地语分词器和26,000小时真实对话数据训练，提出了首个开放、可复现的印地语全双工口语对话系统，实现了自然的打断、重叠和反馈行为。
摘要：
全双工口语对话系统能够模拟自然的对话行为，如打断、重叠和反馈，然而这类系统在印度语言中仍 largely unexplored。我们通过适配最先进的双工语音架构Moshi，使用自定义印地语分词器，并在从14,695名说话者收集的26,000小时真实自发对话数据（具有独立的说话者通道）上进行训练，提出了首个开放、可复现的印地语全双工口语对话系统，从而能够直接从自然交互中学习话轮转换和重叠模式。为了支持印地语文本生成，我们替换了原始英语分词器，并重新初始化了依赖于文本词汇的参数，同时保留了预训练的音频组件。我们提出了一种两阶段训练方案——大规模预训练，然后在1,000小时对话数据上进行微调。通过提示对话延续范式，结合自动评估指标和人工判断，评估结果表明生成的模型在印地语中表现出自然且有意义的全双工对话行为。这项工作为印地语及其他印度语言的实时双工口语对话系统迈出了第一步。
英文摘要：
Full-duplex spoken dialogue systems can model natural conversational behaviours such as interruptions, overlaps, and backchannels, yet such systems remain largely unexplored for Indian languages. We present the first open, reproducible full-duplex spoken dialogue system for Hindi by adapting Moshi, a state-of-the-art duplex speech architecture, using a custom Hindi tokeniser and training on 26,000 hours of real spontaneous conversations collected from 14,695 speakers with separate speaker channels, enabling direct learning of turn-taking and overlap patterns from natural interactions. To support Hindi text generation, we replace the original English tokeniser and reinitialise text-vocabulary-dependent parameters while retaining the pre-trained audio components. We propose a two-stage training recipe -- large-scale pre-training followed by fine-tuning on 1,000 hours of conversational data. Evaluation through the prompted dialogue continuation paradigm with both automatic metrics and human judgments demonstrates that the resulting model generates natural and meaningful full-duplex conversational behaviour in Hindi. This work serves as a first step toward real-time duplex spoken dialogue systems for Hindi and other Indian languages.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
