It allows you to explore audio supported flashcards.

让你能查阅语音支持的卡片进行记忆。

And the long-sought holy grail of ubiquitous voice recognition really will eventually happen.

长期以来向往无所不在的语音识别最终将成为现实。

And developing guardrails for the toys to keep voice interaction safe doesn't come cheap.

而为玩具开发保护语音全的防护措施并不便宜。

Calls have been made in Spanish, Yiddish, Mandarin, Cantonese and Haitian Creole.

已经生成了西班牙语、意第绪语、普通话、广东话和海地克里奥尔语的自助语音电话。

Big advances in speech recognition technology have enabled 8mmunication tools to turn live speech into text in real time on your smartphone or laptop, and sign language into text as well.

语音识别技术的巨大进步使通工具能够在智能手机或笔记本电脑上将实时语音转换为文本，也可以将手语转换为文本。

Built-in voice recognition and Wi-Fi connectivity bring internet searches to life, Microsoft says it will change the future of work.

内置的语音识别和Wi-Fi连接将联网搜索引入了生活，微软表示，这将改工作的未来。

Recognising speech is, however, only the first part of translation.

然而语音识别只是翻译的第一步。

Robocalls are made using what's called voice over internet protocol or VOIP for short.

机器电话使用的是所谓的语音通过联网协议或简称 VOIP。

Confirm you are receiving the Russian voice message.

确认你正在接收俄罗斯的语音息。

Like, you know, the - - the voice assistance, it's all about learning and trying to be better, to serve you better as opposed to just being a blank, plastic, you know, display.

就像，你知道的，语音辅助，这都是关于学习和努力得更好，更好地为你服务，而不只是一个空白的，塑料的，你懂得，像展示一样。

Additionally, interactive hologram televisions will exist, meaning that you can verbally call up your hologram TV and it will appear right in front of you.

此外，交式全息电视将会存在，这意味着你可以用语音呼叫你的全息电视，它就会出现在你面前。

We'll just have to doall the speech recognition work on the Windows computers.

我只需要在Windows电脑上进行所有的语音识别工作。

And basically, it's a text to voice machine.

它基本上就是是一个将文本转换成语音的机器。

And we spent thousands of hours reimagining and redesigning new types of compute interactions, ranging from complex voice commands to intricate hand gestures, all in service of trying to find more natural ways to interact with compute.

我花了上万个小时重新想象、重新设计新型的计算交方式，从复杂的语音控制到精细的手势，都是为了寻找与计算机技术更自然的方式。

Google already has a voice and vedio chat service. So their reasoning is less clear.

谷歌现已有语音和视频聊天的功能。因此，谷歌收购skype的意图并不明确。

That makes 18, including 3 hours in the language lab.

一共18个小时，包括在语音室的3个小时。

And that's the point of course-Google doesn't want Duplex to sound like a machine.

当然这也是谷歌语音助手的特点，谷歌不希望Duplex听起来像机器。

Instead, many opt to call different forms of speech “varieties.”

取而代之，很多选择称其为不同“异”的语音形式。

But machine learning solved a similar problem in human speech recognition.

但机器学习解决了类语音识别中的类似问题。

He's helping fellow veterans share their stories visually and vocally.

他正在帮助退伍军以视觉和语音方式分享他的故事。