Gracias por tu visita. Esta página solo está disponible en inglés.

AVS Documentation Changelog


New API documentation has been added to support the Alexa Presentation Language (APL), which extends the AVS to enable multi-modal experiences on Alexa Built-In devices. See Alexa Multi-Modal API Overview for a summary.

Section 7 of the AVS Functional Requirements has been updated to include devices using APL to render Alexa visual responses.


The AVS Settings paradigm has changed in favor of individually named messages for each setting. The Per-Interface Settings page describes the new approach baed on whether changes are initiated by the cloud or by the device, as well as validating whether settings are synchronized. The changed paradigm includes being able to report a state and values for settings like locale and time zone.

The System Interface page has been updated to include version 2.0, and the changes are listed on the page. As part of this change, the Settings interface has been deprecated. In addition, the Alexa.ApiGateway interface replaces the SetEndpoint functionality.

In coordination with the Settings paradigm described above, there is an additional settings-related functional requirement 8.3 that focus on allowing the customer to choose their preferred languages.

In addition, Alexa has a new Multilingual Mode which, if you use it in your product, must be supported in the companion app settings. Using multilingual mode, Alexa is able to understand customer utterances in multiple languages and reply in the language the customer is currently using. Details can be found on the updated System Interface page.


The UX Design section was updated to include these changes:


The Attention System UX page now includes the following guidance:

  • Text displaying the attention state should not be personalized or presented in the first person.
  • Touch-initiated products must use the Listening start and end audio cues.
  • Linear voice chrome is preferred over circular voice chrome for devices with screens.
  • Circular voice chrome examples have been added.


The default base URLs for Alexa Voice Service connections have changed. As of May 22nd, 2019, the default base URLs for AVS have changed. We recommend that all new and existing clients adopt these new URLs; however, the legacy base URLs will continue to be supported.

Base URLs

Region Supported Countries/Regions URL
Asia Australia, Japan, New Zealand
Europe Austria, France, Germany, India, Italy, Spain, United Kingdom
North America Canada, Mexico, United States

Legacy Base URLs

Region Supported Countries/Regions URL
Asia Australia, Japan, New Zealand
Europe Austria, France, Germany, India, Italy, Spain, United Kingdom
North America Canada, Mexico, United States


A new page has been added presenting attention system design guidance for calling and drop-in, messaging, and announcements on AVS products which include those functions. The Attention System - Calling, Messaging, and Announcements page includes behavior descriptions as well as visual and sound cues.


To help tune products for Echo Spatial Perception, and to give more constant noise suppression performance across products, we have added AVS Functional Requirement 2.5: "The microphones used for Alexa interactions on your product SHALL have +/- 1 dB sensitivity matching."


  • A new design guidance page has been added describing the behavior, location, and appearance of buttons for Alexa Built-in devices.
  • A new design guidance page has been added to suggest how Alexa Built-in devices should handle interrupt behavior and prioritizing customer interactions.
  • The AVS Functional Requirements have been updated to reflect the new button and interrupt guidance:
    • Added requirements relating to the Action button (1.3) and Alerts (5).
    • Clarified requirements for wake word and interrupt guidance in voice-initiated products (2) and touch-initiated product (3.4).
  • In the AVS UX Logo and Brand Usage page, the color called "black" was actually "squid ink". The color codes supplied were correct, but were referred to incorrectly. The color name has been updated.


Alexa is now available in Canadian French. fr-CA is now a supported locale for Settings.SettingsUpdated events.


  • Clarified the UX guidance for visual and sound cues when Do Not Disturb is toggled on and off. In addition, DND information was added to the Multicolor Linear and Multicolor Single tabs of the LED example table.
  • The recommended UX behavior of the Reminder Short response in the LED example table was expanded to cover what should happen when the device is unable to play a speech response. In addition, the sound cue was added to the table and can be stored on device as a fall-back.


  • Opus is now a supported format for captured audio. Review the specification under the Recognize event.
  • DoNotDisturb interface is now available for AVS. Declaring this interface via the Capabilities API will enable users to block incoming notifications, announcements and calls for your Alexa Built-in product. This interface contains one new directive, SetDoNotDisturb, and two events, ReportDoNotDisturb and DoNotDisturbChangedEvent.
  • Do Not Disturb UX guidance has been added to the AVS UX Attention System document, including information on enabling/disabling behavior and visual and sound cues.
  • Do Not Disturb LED animation has been added to the AVS Resource Files for all arrays.


  • AVS is now available in Mexico. Read the blog.
  • The AVS Hosted Splash Screen is now available. See our announcement on the Alexa Blogs. You'll need to request the new alexa:voice_service:preauth scope and the alexa:all scope when using Login with Amazon (LWA) to register a new customer (and remove your existing splash screen if applicable). The AVS Hosted Splash screen will then be deployed automatically to the device that creates a LWA request in the companion app. See our reference documentation and UX guidance for more details.
  • Display cards are now required for TVs, Tablets, and Set-top boxes. This change is reflected in Section 7 of the Functional Requirements.



  • We have updated the Alexa attention state colors to more closely align with Amazon devices in order to provide customers with as much consistency in the Alexa experience as we can. Specifically, we have updated Screen Hex values for Red, Green, Purple, and Orange. The LED Hex values remain the same.


  • We've updated the font resources in the Developer Portal for you to use according to the UX design guidelines. We're updating Amazon Ember to Amazon Ember Display, and we're adding Bookerly (which is used mainly for Things to Try text) and Noto Sans CJK for Chinese, Japanese, and Korean language support.


  • The InteractionModel interface is now available for AVS. This enables you to support the Alexa Routines feature in your products. To enable Alexa Routines, you must declare the InteractionModel 1.0 in the Capabilities API.
  • We have updated the Interaction Model voice request lifecycle to account for the addition of the optional InteractionModel interface.
  • We have updated the TemplateRuntime interface. The RenderPlayerInfo directive has been expanded to support Audible and Flash Briefing, and the controls list includes new media control events. TemplateRuntime remains at version 1.0 in this release.
  • The PlaybackController interface has been upgraded to v1.1, and includes two new events - ButtonCommandIssued and ToggleCommandIssued. These events enable your product to support new controls (skip forward, skip backward, shuffle, repeat, loop, thumbs up, and thumbs down) for media playback.
  • We have updated Display Card UX guidance (Tablet and TV) for playbackController 1.1 and modified the example images. We added a new resource link section at the bottom of each page
  • We have included a new UX guidance page for low-resolution displays.



Find the Ember font files to use for your text display on the Amazon Developer Portal under the Resources tab.



  • The EqualizerController interface allows a product to adjust equalizer settings using Alexa, such as decibel (dB) levels and modes.
  • The InputController interface enables selecting and switching inputs on an Alexa Built-in product.







  • Released the Supported Countries API, which allows a client to retrieve a list of countries or regions where Alexa is available.




  • Clarifying edits have been made to the Interaction Model to emphasize that dialogRequestIds must be unique to each Recognize event, and must not be reused during a session. See Voice Request Lifecycle.



  • NEW StreamingStarted and StreamingEnded events were added to the Bluetooth interface. These events are sent to notify Alexa of streaming state changes. See Manage Bluetooth Streaming State for details.
  • Updated parameter descriptions for the Recognize event and ExpectSpeech directive.


  • AVS is now available in Australia and New Zealand. Read more »


  • AVS is introducing two changes that may require updates to your product’s client code: 1) Cloud-based wake word verification is now required for all voice-initiated products; 2) the initiator type in ExpectSpeech directives has changed from a string to an object. Read more »


Additional guidance added to documentation for the SetAlert directive and AlertStopped event.



  • NEW Developer documentation is now available in Japanese. Language can be toggled within the Developer portal, bottom of page.




  • Updated Alerts documentation updated to reflect support of Named Timers and Reminders. See Alerts interface.




  • New - [Enable Cloud-Based Wake Word Verification][avs-cloud-based-wake-word-verification]
  • New - Streaming Requirements for Cloud-Based Wake Word Verification
  • Updated - SpeechRecognizer Interface
    • The Recognize event and ExpectSpeech directive have been updated to support cloud-based wake word verification.
  • Updated - PlaybackController Interface
    • Sample messages have been updated to include SpeechRecognizer.RecognizerState in the context object. This is only required for clients that use cloud-based wake word verification.
  • Updated - System Interface
    • Sample messages have been updated to include SpeechRecognizer.RecognizerState in the context object. This is only required for clients that use cloud-based wake word verification.
  • Updated - Context
    • SpeechRecognizer.RecognizerState sample message added.
  • Updated - Managing an HTTP/2 Connection
    • Sample messages have been updated to include SpeechRecognizer.RecognizerState in the context object. This is only required for clients that use cloud-based wake word verification.



  • LWA Mobile SDK: The Login with Amazon (LWA) Mobile SDKs for Android and iOS were updated in November 2016. If you are using a companion app or authorizing directly from a product, you’ll need to integrate the latest version of the LWA Mobile SDK to support AVS’s international features. See Prepare for Internationalization.
  • EU Endpoint: A new AVS endpoint for Europe is available that provides customers in the UK and Germany with access to their personal data and country or region specific content.
  • SetEndpoint Directive: The SetEndpoint directive instructs your client to switch endpoints if a user’s settings are not supported on the endpoint your product is connected to.
  • Settings Interface: The new SettingsUpdated event allows a client to adjust Alexa settings, such as locale, using on-product controls or a companion app.


  • Automatic Speech Recognition (ASR) Profiles - AVS provides a choice of three ASR Profiles tuned for different products, form factors, acoustic environments and use cases: CLOSE_TALK, NEAR_FIELD, and FAR_FIELD. For additional information, see Audio Hardware Configurations.
  • Cloud endpointing is now supported, and allows Alexa to determine when a user has stopped speaking. Previously only client endpointing was supported. This new feature is enabled by default when using the NEAR_FIELD or FAR_FIELD ASR Profiles. To learn more, see the Recognize event.
  • The StopCapture directive is designed for cloud endpointing. When Alexa determines that a user has stopped speaking, the StopCapture is sent from Alexa to your client instructing it to stop listening.


  • The Alexa Voice Service (AVS) API was versioned from v1 to v20160207. This release introduces the following capabilities:
  • Volume Control: Ability to adjust and mute/unmute volume on your product using voice or buttons. For more information, see Speaker interface.
  • Media Control: Ability to control media playback (e.g. play/pause) on your product using voice or buttons. For more information, see AudioPlayer interface and PlaybackController interface.
  • Timers and Alarms: Ability to create and manage timers/alarms on your product using voice. For more information, see Alerts interface.
  • Amazon Alexa App Enhancements: Control volume, media playback (e.g. Amazon Music, Audible), as well as manage timers/alarms on your product using the Amazon Alexa App.
  • Quick start guide for building the Java Sample App on Raspberry Pi published to GitHub.