This is a .1 bug fix release for only the JavaScript SDK, not changes are made to the other speech SDKs or SPX command line.

Bug fixes
○ JS - Fix for "audio input reattach fix (#401 and #402 )"
○ JS - Fix for "translation add ability to update target languages (#404)"
○ JS - Fix for "xmlhttprequest-ts dependency with bent, unify REST calls. fixes #403 (#408)
○ JS - Fix for "Fix for audio regression, and add test (#410)"
○ JS - Fix for "Race condition where audio context is closed twice on FireFox (#412)"
JS - Fix for "[AudioConfig]:remove debug console error (#413)"

New features

• set default synthesis voices to neural voice (#376)
• Use AudioWorkletNode by default for mic input (#397)
• getActivationPhrasesAsync API added to VoiceProfileClient class for receiving a list of valid activation phrases in speaker recognition enrollment phase for independent recognition scenarios.
• VoiceProfileClient's enrollProfileAsync API is now async awaitable. See this independent identification code for example usage.

Bug fixes

• fix viseme and bookmark tests failure (#393 )
• fix word level json offsets (#396 )
• fix push stream not being reconnected to with start-stop-start continuous recognition, add test (#399)
• Browser microphone audio processing now uses AudioWorkletNode instead of deprecated ScriptProcessorNode (#391)
• Correctly keep conversations alive during long running conversation translation scenarios. (#389)
• Fixed issue with recognizer reconnecting to a mediastream in continuous recognition. (#385)
• Fixed issue with recognizer reconnecting to a pushStream in continuous recognition. (#399)
• Correct word level offset calculation in detailed recognition results. (#394)

New features

• added retry for connection failures that will make your JavaScript based speech applications more robust.

Bug fixes

• Fixed GitHub issue #366 where ConversationTranslator threw an error 'this.cancelSpeech is not a function'.
• Fixed GitHub issue #298 where 'Get result as an in-memory stream' sample played sound out loud.
• Fixed GitHub issue #350 where calling AudioConfig could result in a 'ReferenceError: MediaStream is not defined'.
• Fixed an UnhandledPromiseRejection warning in Node.js for long-running sessions.

New features

• Node.js users can now use the AudioConfig.fromWavFileInput API. This addresses GitHub issue #252.
• Added two new output audio formats with WebM container for TTS (Webm16Khz16BitMonoOpus and Webm24Khz16BitMonoOpus). These are better formats for streaming audio with the Opus codec. documentation
• Added VisemeReceived event for TTS/speech synthesis to return synchronous viseme animation. See documentation here.
• Added BookmarkReached event for TTS. You can set bookmarks in the input SSML and get the audio offsets for each bookmark. See documentation here.

Bug fixes

• Large WAV file headers are now parsed correctly (increases header slice to 512 bytes). This addresses GitHub issue #962.
• Corrected microphone timing issue if mic stream ends before stop recognition, addressing an issue with Speech Recognition not working in Firefox.
• We now correctly handle initialization promise when the browser forces mic off before turnOn completes.
• We replaced url dependency with url-parse. This addresses GitHub issue #264.

Samples

• Browser samples no longer require separate JavaScript library file download.

New features

• 1.15.1 is bug fix release only.

bug fixes

• DialogServiceConnector: Restore support for Direct Line tokens (#318 ) - fixed an issue that prevented assignment of a channel token for channel authentication.

New features

• JavaScript: BotFrameworkConfig now has fromHost() and fromEndpoint() factory methods that simplify the use of custom service locations versus manually setting properties. We also standardized optional specification of botId to use a non-default bot across the configuration factories.
• JavaScript: Add string control property for websocket compression. For performance reasons we disabled websocket compression by default. This can be reenabled for low bandwidth scenarios. More details here. This addresses this GitHub issue.

bug fixes

• DialogServiceConnector did not previously honor the optional botId parameter specified in BotFrameworkConfig's factories. This made it necessary to set the botId query string parameter manually to use a non-default bot. The bug has been corrected and botId values provided to BotFrameworkConfig's factories, including the new fromHost() and fromEndpoint() additions, will be honored and used. This also applies to the applicationId parameter for CustomCommandsConfig.

• Fixed GitHub issue #881, allowing recognizer object re-usage.

• Fixed an issue where the SKD was sending speech.config multiple times in one TTS session, wasting bandwidth.

• Simplified error handling on microphone authorization, allowing more descriptive message to bubble up when user has not allowed microphone input on their browser.

• Fixed GitHub issue #249 where type errors in ConversationTranslator and ConversationTranscriber caused a compilation error for TypeScript users.

New features

• bug fixes only.

Bug fixes

• Keep types on our custom events without forcing them into a CustomEvent instance (#267)
  This fixes a regression where Connection class callbacks would not propagate back correctly.
• Check if Blob is defined before return Blob type (#269 )
  This fixes an issue for Node users whose dependencies include a defined XMLHttpRequest class.

New features

• Added support for the ConversationTranscriber API. Read documentation here.
• Switch SDK to use Native promises. (#198 )

Bug fixes

• Fixed mapping of session IDs to improve logging and aid in internal debug/service correlations.
• Fixed build with Typescript v4 (#236 ), Thanks you @orgads
• Guid: Use uuid (#242 ) (#259 ), Thanks you @orgads
• Address github issue #228, allow umbrellaIntent to be set for continuous recognition (#254 )
• Add support media stream track (#260 ), Thanks @enricop89
• Added fix for DialogServiceConnector disabling ListenOnce calls after the first call is made.
• Fixed issue where result output would only ever be "simple".
• Fixed continuous recognition issue in Safari on macOS.
• CPU load mitigation for high request throughput scenario (#247 ).
• Allow access to details of Voice Profile Enrollment result.
• Added fix for continuous recognition in IntentRecognizer.

Bug Fixes

• Fix wrong import statements (#217) Thank you @bivas6

New features

• JavaScript: Added Speaker Recognition support for both browser and node.js.
• JavaScript: Added support for automatic language detection/language ID. See documentation here.

Bug fixes

• JavaScript: Fixed an issue with throttling when browser is minimized.
• JavaScript: Fixed an issue with a memory leak on streams.
• JavaScript: Added caching for OCSP responses from NodeJS.
• Java: Fixed an issue that was causing BigInteger fields to always return 0.

New features

• Support speech synthesis (text-to-speech) in NodeJS. Learn more here.
• Add new API's to enable inspection of all send and received messages. Learn more here.

Bug fixes

• Added support for Regions in China with the fromSubscription API. Details here.
• Add more error information for connection failures from NodeJS.

Bug fix release only.

Bug Fixes:

• Something in the react tool chain became unhappy (#156 )

New Features:

• (Preview) Multi-device Conversation: added support for JavaScript and Node.js
• (Preview) Multi-device Conversation: added support for getting a Connection from a ConversationTranslator.
  • You can use this to register for connected/disconnected events
  • You can use this to reconnect to an existing conversation using the Open() function if you get disconnected due to networking or other issues

Samples:

•  Added Multi-device Conversation quickstarts for JavaScript
• Allow setting a custom EndPoint for the Dialog Connector

Big Fixes:

• Enable setting EndpointId's on the SpeechTranslationConfig object.
• Enable selecting a non-default microphone.
• Fix bug in using "process" object that wasn't available in browsers.

Other Changes:

• StopContinuousRecognitionAsync() will now wait to call its continuation callback until any pending audio sent to the Speech Service has been processed. Previously it aborted all pending audio immediately.
  Removed the 10 minute limit for File based Speech Recognition. The Speech SDK will now use the wave header in the file in the event it needs to reconnect to the service.

Speech SDK for JavaScript 1.8.1 is bug fix release and contains the following changes:

• Add header to initial audio send (#126)
• Don't send auth header if there's no auth value (#132)
• Revert back to type script 3.5.3 to resolve compatibility problems. (#133)
• Fix missing null check for Android WebView (#127)

New Features

• JavaScript: Added a feature to Connection object to pass through custom messages from the Speech Service as callback receivedServiceMessage.
• JavaScript: Added support for FromHost API to ease use with on-prem containers and sovereign clouds. See documentation here.
• JavaScript: We now honor NODE_TLS_REJECT_UNAUTHORIZED thanks to a contribution from orgads. See details here.

Bug fixes

• JavaScript: Fixed a circular import of audio data thanks to a contribution from euirim.
• JavaScript: added support for setting service properties, as added in 1.7.
• JavaScript: fixed an issue where a connection error could result in continuous, unsuccessful websocket reconnect attempts.

Speech SDK for JavaScript 1.8.1 contains the following changes:

• [Stability] Address exception coming from deref attempt of null streams. (#104 )
• [Minor rename] Update to the latest dialog type, CustomCommands. (#107 )
• Adds support for the “Conversation_From_Id” property, previously this was not being passed with the AgentConfig and was thus ignored. Add support for requesting detailed response info, you can now specify with SpeechServiceResponse_OutputFormatOption. Resolves the issue where a new reco turn starting before the previous one had finished would cause the audio turn manager state to become corrupt. Also added a flag to ensure we don’t start a new reco turn before the previous finishes. (#110 )

Speech SDK for JavaScript 1.8.0 contains the following changes:

• Don't reuse variable names in telemetry verification (#89)
• Process speech.fragment message as a speech.hypothesis (#87)
• Update properties & URL parameter support on config objects. (#88)

Also these upkeep and dependency updates:

• Fix machine in the middle vulnerability for "https-proxy-agent" (#97)
• Update transitive dependencies to resolve security warnings (#84)

Speech SDK for JavaScript 1.7.0 contains the following changes:

• Update Dependencies [Thanks to @orgads #77 ]
• Align SpeechConfig properties with native SDK #88
• Improve error handling when subscription quota is exceeded #83

Speech SDK for JavaScript 1.6.0 contains the following changes:

• Remove throttling for the first 5 seconds of audio. (#68)
• Fixup documentation links to render properly in TypeDoc. (#69)
• Add OCSP certificate validation when connecting from node.js (#70)
• Provide es2015 module. (#52) and Ignore ocsp for es2015 build (#71) - Thank you @Domratchev.

Speech SDK for JavaScript 1.5.0 contains the following changes:

• Add demo bundle project (#50)
• Fix version information on proxy properties and functions (#54)
• Updated static analysis and build tools to resolve security warnings in their dependency chain. (Security updates #53 and #61)
• Fix incorrect translation offsets & latency telemetry (#63)