最近调研了一下Amazon Alexa智能语音服务，这里记录一下调研到的信息；
Alexa基本上就是Amazon智能音箱Echo的大脑，比如用查天气举例：用户说“What's the weather in Seattle”，Echo都会发送至Alexa去处理，提取特征“weather”、“Seattle”，分析用户的语义，再根据语义去查天气信息，然后将查到的信息返回给Echo，Echo收到信息之后播报。
Alexa
Alexa, the voice service that powers Echo, provides capabilities, or skills, that enable customers to interact with devices in a more intuitive way using voice. Examples of these skills include the ability to play music, answer general questions, set an alarm or timer and more. Alexa is built in the cloud, so it is always getting smarter. The more customers use Alexa, the more she adapts to speech patterns, vocabulary, and personal preferences.
Echo的智能语音服务，为用户提供语音交互服务，例如播放音乐、回答问题、设置闹钟等
支持的设备
- Amazon Echo等
- 集成了AVS（Alexa Voice Service）的第三方厂家设备
- 手机应用（集成AVS，通过http2的方式调用AVS API）
AVS（Alexa Voice Service）
Use the Alexa Voice Service (AVS) to add intelligent voice control to any connected product that has a microphone and speaker. Your customers will be able to ask Alexa to play music, answer questions, get news and local information, control smart home products, and more on their voice-enabled products.
AVS is now available for the UK and Germany.目前支持英文和德语
AVS provides hardware and software development tools to help you easily build products with Alexa.
Demo地址
这个Demo的思路比较奇怪，比如没有用Android直接集成AVS，而是先在Mac上跑一个Java程序（自带图形界面），然后打开Android模拟器，运行它的android程序，android程序获取授权得到Token，然后再使用Mac的那个java程序来测试。（个人觉得这逻辑很怪...为什么不直接通过http2的方式调用AVS API...）
这里再强调一下，AVS不支持中文，所以Demo测试时要用英文
一个同事在Mac尝试了该Demo，Demo跑起来之后，可以在Mac上说话，比如你说“What's the weather in Seattle”,Demo会回复“天气是xxxxxx”
Demo的坑
Android获取授权那一步，最好用原生的模拟器，不要用Genymotion的模拟器，否则会连接超时；
Building with AVS
Integrate Alexa into your next voice-enabled connected product. Alexa is always getting smarter with new capabilities and services through machine learning, regular feature updates, and custom skills. The AVS speech engines and features live in the cloud and are easily accessible via APIs, minimizing the design complexity of your product. We also provide software and hardware development tools to help you quickly and easily build Alexa-enabled products.
可通过AVS Device SDK、Development Kits for AVS、AVS API三种方式集成AVS
一个android集成AVS的例子
AVS Device SDK
The AVS Device SDK provides C++-based libraries that enable your device to process audio inputs and triggers, establish persistent connections with AVS, and handle all Alexa driven interactions. The SDK leverages the AVS API that handle core Alexa functionality, including speech recognition and synthesis, and other capabilities such as streaming media, timers and alarms, notifications, weather reports, and thousands of custom skills.
Development Kits for AVS
Development Kits for AVS are reference solutions for integrating Alexa into your products. They include chipsets, voice processing technologies, and client software that leverages the AVS APIs to help you easily build voice-enabled commercial products while reducing development costs and accelerating the integration process. These solutions range in features, cost, and in use case, allowing you to select the best performing audio front end solution for your product needs.
AVS API
AVS provides programming interfaces to access the cloud-based service from your product, regardless of its operating system. Each interface contains logically grouped messages called directives and events that enable your device and the AVS cloud to communicate directly. The APIs handle core Alexa functionality, including speech recognition and synthesis, and other capabilities such as streaming music, timers and alarms, notifications, weather reports, and skills—enabling your device to capture an utterance, process it in the cloud, and playback an intelligent speech response.
What Is an Alexa Skill
Alexa is Amazon’s voice service and the brain behind millions of devices like the Amazon Echo, Echo Dot, and Echo Show. Alexa provides capabilities, or skills, that enable customers to create a more personalized experience. There are now more than 15,000 skills from companies like Starbucks, Uber, and Capital One as well as other innovative designers and developers.
What Is the Alexa Skills Kit?
With the Alexa Skills Kit (ASK), designers, developers, and brands can build engaging skills and reach millions of customers. ASK is a collection of self-service APIs, tools, documentation, and code samples that makes it fast and easy for you to add skills to Alexa. With ASK, you can leverage Amazon’s knowledge and pioneering work in the field of voice design.
Alexa Skills Kit用于扩展Alexa能力，第三方（开发者、服务商等）使用ASK开发应用，来提供查找信息、订餐、订车、控制智能家居设备、获取新闻资讯等功能。
skill类型
-
Custom Skills
- Look up tide information
- Order a pizza
- Request a taxi
- Engage the user in a game, such as word puzzles or trivia
- Just about any other action you can imagine!
-
Smart Home Skills
- turn on / turn off
- increase / decrease the temperature
- change the dimness or brightness for a light
you (as the developer) define: How your skill responds to a particular directive. For instance, you write the code that makes a light turn on when your skill receives a “turn on the light” directive. This code is called a skill adapter.
Flash Briefing Skills
For this type of skill, the Flash Briefing Skill API defines:
-
The words users say to make (or invoke) those requests. For example:
- “give me my flash briefing”
- “tell me the news”
-
You (as the creator) define:
- The name, description and images for a flash briefing skill. This helps a customer choose your skill in the skill store.
- One or more content feeds for a flash briefing skill. These feeds can contain audio content that is played to the customer or text content that Alexa reads to the customer.
-
Video Skills
A video skill enables you to provide video content such as TV shows and movies for customers.
- The words users say to make (or invoke) those requests. For example:
- “play Manchester by the Sea”
- “change to channel 4”
You (as the creator) define:
The name, description and images for a video skill. This helps a customer choose your skill in the skill store.
The requests the skill can handle such as playing and searching for video content and how video content search results display.
- The words users say to make (or invoke) those requests. For example:
案例
- 接入亚马逊Echo音箱教程
- 一文看懂 Echo 和 Alexa，亚马逊如何用苹果的玩法在玩语音？
