发行说明:语音设备 SDKRelease notes: Speech Devices SDK

以下部分列出了最新版本中的更改。The following sections list changes in the most recent releases.

语音设备 SDK 1.11.0:Speech Devices SDK 1.11.0:

语音设备 SDK 1.9.0:Speech Devices SDK 1.9.0:

  • 提供了 Urbetter DDK 的初始二进制文件 (Linux ARM64)。Initial binaries for Urbetter DDK (Linux ARM64) are provided.
  • Roobo v1 现在使用适用于语音 SDK 的 MavenRoobo v1 now uses Maven for the Speech SDK
  • 已将语音 SDK 组件更新到 1.9.0 版。Updated the Speech SDK component to version 1.9.0. 有关详细信息,请参阅其发行说明For more information, see its release notes.

语音设备 SDK 1.7.0:Speech Devices SDK 1.7.0:

  • 现在支持 Linux ARM。Linux ARM is now supported.
  • 提供了 Roobo v2 DDK 的初始二进制文件 (Linux ARM64)。Initial binaries for Roobo v2 DDK are provided (Linux ARM64).
  • Windows 用户可以使用 AudioConfig.fromDefaultMicrophoneInput()AudioConfig.fromMicrophoneInput(deviceName) 指定要使用的麦克风。Windows users can use AudioConfig.fromDefaultMicrophoneInput() or AudioConfig.fromMicrophoneInput(deviceName) to specify the microphone to be used.
  • 库大小已经过优化。The library size has been optimized.
  • 支持使用相同的语音/意向识别器对象进行多轮次识别。Support for multi-turn recognition using the same speech/intent recognizer object.
  • 解决了偶尔出现的问题,即进程会在停止识别时停止响应。Fix occasional issue where the process would stop responding while stopping recognition.
  • 示例应用现在包含一个示例 participants.properties 文件,用于演示文件的格式。Sample apps now contain a sample participants.properties file to demonstrate the format of the file.
  • 已将语音 SDK 组件更新到 1.7.0 版。Updated the Speech SDK component to version 1.7.0. 有关详细信息,请参阅其发行说明For more information, see its release notes.

语音设备 SDK 1.6.0:Speech Devices SDK 1.6.0:

语音设备 SDK 1.5.1:Speech Devices SDK 1.5.1:

语音设备 SDK 1.5.0:2019 年 5 月发布Speech Devices SDK 1.5.0: 2019-May release

  • 语音设备 SDK 现为 GA,不再是受限的预览版。Speech Devices SDK is now GA and no longer a gated preview.
  • 已将语音 SDK 组件更新到 1.5.0 版。Updated the Speech SDK component to version 1.5.0. 有关详细信息,请参阅其发行说明For more information, see its release notes.
  • 新的关键字技术带来了显著的质量改进,请参阅“中断性变更”。New keyword technology brings significant quality improvements, see Breaking Changes.
  • 新的音频处理管道改进了远端域识别。New audio processing pipeline for improved far-field recognition.

重大更改Breaking changes

  • 由于新出现的关键字技术,所有关键字必须在改进的关键字门户中重新创建。Due to the new keyword technology all keywords must be re-created at our improved keyword portal. 若要从设备中完全删除旧的关键字,请卸载旧应用。To fully remove old keywords from the device uninstall the old app.
    • adb uninstall com.microsoft.cognitiveservices.speech.samples.sdsdkstarterappadb uninstall com.microsoft.cognitiveservices.speech.samples.sdsdkstarterapp

语音设备 SDK 1.4.0:2019 年 4 月发布Speech Devices SDK 1.4.0: 2019-Apr release

语音设备 SDK 1.3.1:2019 年 3 月发布Speech Devices SDK 1.3.1: 2019-Mar release

  • 已将语音 SDK 组件更新到 1.3.1 版。Updated the Speech SDK component to version 1.3.1. 有关详细信息,请参阅其发行说明For more information, see its release notes.
  • 更新了关键字处理,请参阅“中断性变更”。Updated keyword handling, see Breaking Changes.
  • 示例应用程序添加了选择语言的功能,适用于语音识别和翻译。Sample application adds choice of language for both speech recognition and translation.

语音设备 SDK 1.1.0:2018 年 11 月版本Speech Devices SDK 1.1.0: 2018-Nov release

  • 已将语音 SDK 组件更新到 1.1.0 版。Updated the Speech SDK component to version 1.1.0. 有关详细信息,请参阅其发行说明For more information, see its release notes.
  • 远场语音识别准确性已通过我们增强的音频处理算法得到提高。Far Field Speech recognition accuracy has been improved with our enhanced audio processing algorithm.
  • 示例应用程序添加了中文语音识别支持。Sample application added Chinese speech recognition support.

语音设备 SDK 1.0.1:2018 年 10 月版本Speech Devices SDK 1.0.1: 2018-Oct release

  • 语音 SDK 组件更新到了 1.0.1 版。Updated the Speech SDK component to version 1.0.1. 有关详细信息,请参阅其发行说明For more information, see its release notes.
  • 我们改进的音频处理算法将提高语音识别的准确性Speech recognition accuracy will be improved with our improved audio processing algorithm
  • 修复了一个连续识别音频会话 bug。One continuous recognition audio session bug is fixed.

重大更改Breaking changes

  • 该版本中推出了大量重大更改。With this release a number of breaking changes are introduced. 有关 API 的详细信息,请查看此页Please check this page for details relating to the APIs.

语音设备 SDK 0.5.0:2018 年 8 月版本Speech Devices SDK 0.5.0: 2018-Aug release

  • 通过修复音频处理代码中的 Bug,改进了语音识别的准确性。Improved the accuracy of speech recognition by fixing a bug in the audio processing code.
  • 语音 SDK 组件更新到了 0.5.0 版。Updated the Speech SDK component to version 0.5.0. 有关详细信息,请参阅其发行说明For more information, see its release notes.

语音设备 SDK 0.2.12733:2018 年 5 月版本Speech Devices SDK 0.2.12733: 2018-May release

认知服务语音设备 SDK 的第一个公共预览版本。The first public preview release of the Cognitive Services Speech Devices SDK.