Skip to content

Latest commit

 

History

History

text-to-speech

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
Assets
Assets
 
 
ProjectSettings
ProjectSettings
 
 
README.md
README.md
 
 

README.md

Quickstart: Synthesize speech in Unity using the Speech SDK for Unity

This sample demonstrates how to synthesize speech in Unity.

Note: The Speech SDK for Unity supports Windows Desktop (x86 and x64) or Universal Windows Platform (x86, x64, ARM/ARM64), Android (x86, ARM32/64), iOS (x64 simulator, ARM64), Mac (x64) and Linux (x64).

Prerequisites

Download the Speech SDK for Unity and the sample code

Prepare the sample

  • Open the sample project in Unity:
    • Start Unity. In the launch window, under the Projects tab select Open.
    • Select the folder which contains this sample.
    • After a bit of time, the Unity Editor window should pop up.

      Note: do not worry about the error The type or namespace name 'CognitiveServices' does not exist in the namespace 'Microsoft' (are you missing an assembly reference?) after opening the sample. This error gets resolved in the next step by importing the Speech SDK.

  • Import the Speech SDK:
    • Import the Speech SDK by selecting Assets > Import Package > Custom Package.
    • In the file picker, select the Speech SDK .unitypackage file that you downloaded before.
    • Ensure that all files are selected and click Import.
    • NOTE: If you have installed Unity only for specific platform e.g. Unity CloudBuild for Android, be aware that you may need to remove other
    • platform (iOS, Mac, WSA) binaries from the Speech SDK Plugins folder to avoid possible conflicts with the same library name.
  • Select the HelloWorld sample scene:
    • In the Project Window (bottom left by default), navigate to Assets > Scenes and double-click on the HelloWorld scene to open it up.
  • Update the sample code with your subscription data:
    • In the Project Window, navigate to Assets > Scripts and double-click the HelloWorld C# script to edit it. Note: you can configure which code editor will be launched under Edit > Preferences, for details see here.
    • Do the following edits:
      • Replace the string YourSubscriptionKey with your own subscription key.
      • Replace the string YourServiceRegion with the service region of your subscription. For example, replace with westus if you are using the 30-day free trial subscription.
      • Save the file.

Note: the HelloWorld C# script is used as a component of the canvas object in the HelloWorld scene.

Run the sample

There are various options to run this sample. Start by running it in the Unity Editor (first option), then explore other options depending on your needs.

Run the sample in the Unity Editor

  • Press the Play button in the Unity Editor toolbar (below the menu bar).
    • In the Unity Editor's Game Window, type into the textbox some text that you want to synthesize. The text is transmitted to the Speech service and synthesized to speech, which will playback on your speaker.
    • Check also the Console Window for debug messages.
  • Click the Play button again to stop running the app.

Build and run the sample as a stand-alone desktop application

  • Open File > Build Settings.
  • Select PC, Mac & Linux Standalone as platform.
  • If this wasn't the active platform before, you have to select Switch Platform (and wait a bit).
  • Ensure the Target Platform field is set to Windows and pick the option you need from the Architecture field.
  • Select Build.
  • In the folder picker, create and select a new build folder.
  • Once the build has finished, the Windows Explorer launches with the folder containing the application HelloWorld.exe. Do not launch it yet
  • Navigate to the HelloWorld_Data/Plugins folder in your build output folder. Copy the DLL Microsoft.CognitiveServices.Speech.core.dll into the HelloWorld_Data/Managed folder.
  • Double-click on HelloWorld.exe in the build output folder to start the application.
  • Then, type text into the textbox and click button. The text is transmitted to the Speech service and synthesized to speech, which will playback on your speaker.

Build and run the sample as Universal Windows Platform application

  • Open File > Build Settings.
  • Select Universal Windows Platform as platform.
  • If this wasn't the active platform before, you have to select Switch Platform (and wait a bit).
  • Ensure that Target Platform and Architecture are set appropriately.
  • Click Player Settings... to open them up in the Inspector Window (on the right by default).
  • Navigate to the section Other Settings in the Settings for Universal Windows Platform tab:
    • Select .NET 4.x Equivalent as Scripting Runtime Version.
    • Select IL2CPP as Scripting Backend.
  • In the section Publishing Settings, review that checkmarks are set for the these capabilities in the Capabilities control:
    • InternetClient
    • InternetClientServer
  • Back in the Build Settings window, click Build And Run.
    • For Windows ARM64 architecture, choose Build and then load the solution file to build for ARM64 and deploy your package to your device. You may hit build error "MSB3273" for the UnityPlayer.dll ImageFileMachine value of "0xAA64". To suppress this you can edit the generated HelloWorld.vcxproj and add "<ResolveAssemblyWarnOrErrorOnTargetArchitectureMismatch>None</ResolveAssemblyWarnOrErrorOnTargetArchitectureMismatch>" to the "Global" property group. For more information see this StackOverflow article.
  • In the folder picker, create and select a new folder.
  • After the build completes, the UWP app launches:
    • Then, type text into the textbox and click button. The text is transmitted to the Speech service and synthesized to speech, which will playback on your speaker.

Build and run the sample for Android platform

  • Complete your Android environment setup in Unity, including the Android NDK setup.
  • Open File > Build Settings.
  • Select Android as platform.
  • If this wasn't the active platform before, you have to select Switch Platform (and wait a bit).
  • Click Player Settings... to open them up in the Inspector Window (on the right by default).
  • Navigate to the Other Settings of the Settings for Android tab.
  • Review the information under Identification and Configuration and make sure they are correct for your environment, in particular:
    • Minimum API Level
    • Target API Level
    • Scripting Runtime Version (set to .NET 4.x)
    • Scripting Backend (set to IL2CPP)
    • Target Architectures
  • Connect your developer-enabled Android device to your PC via USB.
  • Back in the Build Settings window, click Build And Run.
  • After the build completes, the app launches on your Android device:
    • Then, type text into the textbox and click button. The text is transmitted to the Speech service and synthesized to speech, which will playback on your speaker.

Build and run the sample for iOS platform

  • Get started on iOS environment setup in Unity, and prepare iPhone device for development.

  • Open this Unity sample project on Mac.

  • Under Project > Assets > SpeechSDK > Plugins > iOS, check that Microsoft.CognitiveServices.Speech.csharp.dll target is enabled for Editor and iOS platforms.

  • Open File > Build Settings,

  • Select iOS as platform.

  • If this wasn't the active platform before, you have to select Switch Platform (and wait a bit).

  • Click Player Settings... to open them up in the Inspector Window (on the right by default).

  • Check that under Other Settings, Bundle Identifier, Version and Target minimum iOS Version matches with your Apple device provisioning profile.

  • Back in the Build Settings window, click Build and create a new folder where Xcode project will be generated by Unity.

  • After the build has completed, open created Xcode project with your Xcode editor.

  • In Xcode, under Targets > Unity-Phone > General, check that Identity and Signing matches with your provisioning profile.

    Note: Bitcode is enabled in the iOS plugin.

  • Connect your developer-enabled iOS device to your Mac via USB and select that as a target in Xcode.

  • In Xcode, build and run the solution using Play button.

  • After the build completes, the app launches on your iOS device:

    • Then, type text into the textbox and click button. The text is transmitted to the Speech service and synthesized to speech, which will playback on your speaker.

    Build and run the sample for MacOS platform

  • Get started on MacOS environment setup in Unity.

  • Under Project > Assets > SpeechSDK > Plugins > MacOS, check that Microsoft.CognitiveServices.Speech.csharp.dll target is enabled for Editor and Standalone and MacOS platforms.

  • Open File > Build Settings,

  • Select PC, Mac & Linux Standalone as platform.

  • If this wasn't the active platform before, you have to select Switch Platform (and wait a bit).

  • Click Player Settings... to open them up in the Inspector Window (on the right by default).

  • Back in the Build Settings window, click Build and create a new of the standalone app which will be generated by Unity.

  • After the build has completed, open created standalone application and launch it for the speech experience.

    Build and run the sample for Linux platform

  • Get started on Linux environment setup in Unity.

  • Under Project > Assets > SpeechSDK > Plugins > Linux, check that Microsoft.CognitiveServices.Speech.csharp.dll target is enabled for Editor and Standalone and Linux platforms.

  • Open File > Build Settings,

  • Select PC, Mac & Linux Standalone as platform.

  • If this wasn't the active platform before, you have to select Switch Platform (and wait a bit).

  • Back in the Build Settings window, click Build and create a new of the standalone app which will be generated by Unity.

  • After the build has completed, open created standalone application and launch it for the speech experience.

References