倾听、学习与大规模帮助：机器学习如何改变Airbnb的语音支持体验
[
[
A look into how Airbnb uses speech recognition, intent detection, and language models to understand users and assist agents more effectively.
了解 Airbnb 如何使用语音识别、意图检测和语言模型来更有效地理解用户并协助代理。
By Yuanpei Cao, _H_eng Ji, Elaine Liu, Peng Wang, and Tiantian Zhang
作者 Yuanpei Cao、 _H_eng Ji、 Elaine Liu、 Peng Wang 和 Tiantian Zhang
At Airbnb, we aim to provide a smooth, intuitive, and helpful community support experience, whether it’s helping a guest navigate a booking change or helping a host with a listing issue. While our Help Center and customer support chatbot helps resolve many inquiries efficiently, some users prefer the immediacy of a voice conversation with a support representative. To make these interactions faster and more effective, we’ve significantly improved our Interactive Voice Response (IVR) system via machine learning.
在Airbnb，我们旨在提供顺畅、直观和有帮助的社区支持体验，无论是帮助客人处理预订变更，还是帮助房东解决房源问题。虽然我们的帮助中心和客户支持聊天机器人能够高效解决许多咨询，但一些用户更喜欢与支持代表进行语音对话的即时性。为了使这些互动更快、更有效，我们通过机器学习显著改善了我们的互动语音响应（IVR）系统。
Over the years, Airbnb has invested in conversational AI to enhance customer support. In our previous blog posts Task-Oriented Conversational AI in Airbnb Customer Support and Using Chatbots to Provide Faster COVID-19 Community Support, we explored how AI-driven chatbots streamline guest and host interactions through automated messaging. This post explains how we extend that work to voice-based support, leveraging machine learning to improve real-time phone interactions with our intelligent IVR system.
多年来，Airbnb 投资于对话式 AI 以增强客户支持。在我们之前的博客文章 Airbnb 客户支持中的任务导向对话式 AI 和 使用聊天机器人提供更快的 COVID-19 社区支持 中，我们探讨了 AI 驱动的聊天机器人如何通过自动化消息简化客人和房东之间的互动。本文解释了我们如何将这项工作扩展到基于语音的支持，利用机器学习改善与我们智能 IVR 系统的实时电话互动。
We’ll take you through the end-to-end IVR journey, the key machine learning components that power it, and how we designed a syst...