Oh, Jeez! 还是 Uh-huh? 基于 ASR 转录文本的听者感知反向信道预测器
Oh, Jeez! or Uh-huh? A Listener-aware Backchannel Predictor on ASR Transcriptions
AI总结：
本文研究对话中的反向信道建模，提出一种结合词汇、声学线索及听者嵌入的预测器，在Switchboard数据集上验证了声学线索优于词汇线索，且两者结合听者嵌入在手动和自动转录上均效果最佳。
摘要：
本文介绍了我们在对话中反向信道建模的最新研究。受主动反向信道理论启发，我们旨在开发一个系统，通过插入反向信道（如继续词和评估）来影响说话者，充当主动听者。我们的模型不仅考虑词汇和声学线索，还引入了使用听者嵌入来模拟不同反向信道行为的简单新颖想法。我们在 Switchboard 基准数据集上的实验结果表明，声学线索在此任务中比词汇线索更重要，且其与听者嵌入的组合在手动转录和自动生成转录上均表现最佳。
英文摘要：
This paper presents our latest investigation on modeling backchannel in conversations. Motivated by a proactive backchanneling theory, we aim at developing a system which acts as a proactive listener by inserting backchannels, such as continuers and assessment, to influence speakers. Our model takes into account not only lexical and acoustic cues, but also introduces the simple and novel idea of using listener embeddings to mimic different backchanneling behaviours. Our experimental results on the Switchboard benchmark dataset reveal that acoustic cues are more important than lexical cues in this task and their combination with listener embeddings works best on both, manual transcriptions and automatically generated transcriptions.
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。
