Search Unity

RT-Voice - Run-time text-to-speech solution

Discussion in 'Assets and Asset Store' started by Stefan-Laubenberger, Jul 10, 2015.

  1. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hi

    The standalone provider (= offline) won't work with IL2CPP. You will need either MaryTTS or AWS Polly as solution.
    With the new update, Klattersynth is also working offline.

    We're still waiting for a fix from Unity. Please vote on the issue here:
    https://issuetracker.unity3d.com/is...process-dot-start-fails-to-launch-application


    Cheers
    Stefan


    Edit:
    RT-Voice works now with Windows standalone and IL2CPP builds!
     
    Last edited: Apr 11, 2019
    Bartolomeus755 likes this.
  2. Bartolomeus755

    Bartolomeus755

    Joined:
    Jun 20, 2013
    Posts:
    240
    Thx for the info, also voted.
     
    Stefan-Laubenberger likes this.
  3. mavisakal

    mavisakal

    Joined:
    Aug 26, 2017
    Posts:
    15
    Hi @Stefan-Laubenberger ,
    I've created an app for iOS using Salsa3D and RTVoice. Text-to-speech and lipsync works fine in Unity editor. But when I run it in an iOS device the lipsync doesn't work. In your documentation it says iOS can't generate audio files which is needed by Salsa3D . Without lipsync my avatar is useless. I tried the textsync but it's not accurate.
    Are there any other solutions? Can I use a third party TTS service such as AWS Polly to solve this issue? If so can you provide an example?

    thanks
     
  4. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hi @mavisakal

    Yes, you could use MaryTTS, AWS Polly or offline the brand new integration of Klattersynth.
    You can use RTV like before, no code changes needed. Simply enable the desired solution in the Speaker-component.
    There are also manuals under "Documentation" to help you further.

    We also provide an account for testing purposes on our MaryTTS-server - if you want one, please send me the desired username and invoice.

    So long,
    Stefan
     
  5. mavisakal

    mavisakal

    Joined:
    Aug 26, 2017
    Posts:
    15
    Thanks @Stefan-Laubenberger . It worked with AWS Polly!
    But I think I found a bug. For some texts it throws the following error.

    Code (JavaScript):
    1. Could not speak the text: Wrapper {Uid='*********', Text='Man United face Barcelona in UCL as Liverpool draw FC Porto - ESPN. North Korea says it may suspend nuclear talks - CNN. China scrambled to show it'll change how it treats foreign firms — that may not be enough for Trump - CNBC. Tom Clancy video game 'The Division 2' brings a battered Washington, D.C., to life - USA TODAY. Hudson Yards just opened in New York, here's a look inside - CNBC. ', Source='True', Voice='Joanna (en-US, FEMALE)', SpeakImmediately='True', Rate='1', Pitch='1', Volume='1', OutputFile='', ForceSSML='True', Created='3/15/2019 3:35:07 PM'}
    2. Error: Amazon.Runtime.AmazonUnmarshallingException: Error unmarshalling response back from AWS. Request ID: *******, Response Body: Failed to receive data ---> ThirdParty.Json.LitJson.JsonException: Invalid character 'F' in input string
    3.   at ThirdParty.Json.LitJson.Lexer.NextToken () [0x00027] in E:\JenkinsWorkspaces\v3-trebuchet-release\AWSDotNetPublic\sdk\src\Core\ThirdParty\Json\Lexer.cs:884
    4.   at ThirdParty.Json.LitJson.JsonReader.ReadToken () [0x0000a] in E:\JenkinsWorkspaces\v3-trebuchet-release\AWSDotNetPublic\sdk\src\Core\ThirdParty\Json\JsonReader.cs:251
    5.   at ThirdParty.Json.LitJson.JsonReader.Read () [0x0003d] in E:\JenkinsWorkspaces\v3-trebuchet-release\AWSDotNetPublic\sdk\src\Core\ThirdParty\Json\JsonReader.cs:297
    6.   at Amazon.Runtime.Internal.Transform.JsonUnmarshallerContext.Read () [0x0001e] in E:\JenkinsWorkspaces\v3-trebuchet-release\AWSDotNetPublic\sdk\src\Core\Amazon.Runtime\Internal\Transform\JsonUnmarshallerContext.cs:183
    7.   at Amazon.Runtime.Internal.Transform.JsonErrorResponseUnmarshaller.Unmarshall (Amazon.Runtime.Internal.Transform.JsonUnmarshallerContext context) [0x00082] in E:\JenkinsWorkspaces\v3-trebuchet-release\AWSDotNetPublic\sdk\src\Core\Amazon.Runtime\Internal\Transform\JsonErrorResponseUnmarshaller.cs:51
    8.   at Amazon.Polly.Model.Internal.MarshallTransformations.SynthesizeSpeechResponseUnmarshaller.UnmarshallException (Amazon.Runtime.Internal.Transform.JsonUnmarshallerContext context, System.Exception innerException, HttpStatusCode statusCode) [0x00000] in C:\codebase\v3\AWSDotNetPublic\sdk\src\Services\Polly\Generated\Model\Internal\MarshallTransformations\SynthesizeSpeechResponseUnmarshaller.cs:68
    9.   at Amazon.Runtime.Internal.Transform.JsonResponseUnmarshaller.UnmarshallException (Amazon.Runtime.Internal.Transform.UnmarshallerContext input, System.Exception innerException, HttpStatusCode statusCode) [0x00015] in E:\JenkinsWorkspaces\v3-trebuchet-release\AWSDotNetPublic\sdk\src\Core\Amazon.Runtime\Internal\Transform\ResponseUnmarshallers.cs:198
    10.   at Amazon.Runtime.Internal.HttpErrorResponseExceptionHandler.HandleException (IExecutionContext executionContext, Amazon.Runtime.Internal.HttpErrorResponseException exception) [0x00063] in E:\JenkinsWorkspaces\v3-trebuchet-release\AWSDotNetPublic\sdk\src\Core\Amazon.Runtime\Pipeline\ErrorHandler\HttpErrorResponseExceptionHandler.cs:78
    11.   --- End of inner exception stack trace ---
    12.   at Amazon.Runtime.Internal.HttpErrorResponseExceptionHandler.HandleException (IExecutionContext executionContext, Amazon.Runtime.Internal.HttpErrorResponseException exception) [0x000a0] in E:\JenkinsWorkspaces\v3-trebuchet-release\AWSDotNetPublic\sdk\src\Core\Amazon.Runtime\Pipeline\ErrorHandler\HttpErrorResponseExceptionHandler.cs:93
    13. UnityEngine.Debug:LogError(Object)
    14. Crosstales.RTVoice.AWSPolly.<speak>c__AnonStorey6:<>m__0(AmazonServiceResult`2) (at Assets/Plugins/crosstales/RTVoice/3rd party/AWS Polly/Scripts/VoiceProviderAWS.cs:446)
    15. Amazon.Polly.<>c__DisplayClass16_0:<SynthesizeSpeechAsync>b__0(AmazonWebServiceRequest, AmazonWebServiceResponse, Exception, AsyncOptions) (at C:/codebase/v3/AWSDotNetPublic/sdk/src/Services/Polly/Generated/_unity/AmazonPollyClient.cs:346)
    16. Amazon.Runtime.Internal.UnityMainThreadDispatcher:ProcessRequests() (at E:/JenkinsWorkspaces/v3-trebuchet-release/AWSDotNetPublic/sdk/src/Core/Amazon.Runtime/Pipeline/_unity/UnityMainThreadDispatcher.cs:82)
    17. Amazon.Runtime.Internal.UnityMainThreadDispatcher:Update() (at E:/JenkinsWorkspaces/v3-trebuchet-release/AWSDotNetPublic/sdk/src/Core/Amazon.Runtime/Pipeline/_unity/UnityMainThreadDispatcher.cs:59)
     
  6. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hmm, this is a problem in the AWS code. We only included the AWS SDK for convenience and it's in the hands of Amazon.
    I'm not sure if it's already fixed in the latest release (if not, please open an issue on Amazon, but this could take a long time until it's fixed).

    Imho, the text is looking good and it's working inside the Unity Editor.
    Can you please try this: split the text in smaller chunks and "feed" them to AWS Polly until you find the part that's causing the trouble.
     
  7. mavisakal

    mavisakal

    Joined:
    Aug 26, 2017
    Posts:
    15
    Ok I'll check that and will share when done. I have another question though :)
    I would like to implement my custom artificial voice. Lyrebird.ai provides a TTS service for this.
    I want to make a RTVoice wrapper which uses Lyrebird APIs. Just as you do for AWS Polly.
    Which class should I examine to implement this?
     
  8. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Please take a look at the VoiceProviderAWS or VoiceProviderCustomExample.

    I'm very interested in your solution, please keep me updated.
     
  9. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    The customer found a solution!
    Simply add the those lines to a "link.xml"-file inside your project:
    Code (CSharp):
    1. <linker>
    2.     <assembly fullname="UnityEngine">
    3.         <type fullname="UnityEngine.Networking.UnityWebRequest" preserve="all" />
    4.         <type fullname="UnityEngine.Networking.UploadHandlerRaw" preserve="all" />
    5.         <type fullname="UnityEngine.Networking.UploadHandler" preserve="all" />
    6.         <type fullname="UnityEngine.Networking.DownloadHandler" preserve="all" />
    7.         <type fullname="UnityEngine.Networking.DownloadHandlerBuffer" preserve="all" />
    8.     </assembly>
    9. </linker>
    Thank you @mavisakal !
     
    Last edited: Mar 16, 2019
  10. mavisakal

    mavisakal

    Joined:
    Aug 26, 2017
    Posts:
    15
    Hi Stefan,
    It was just the changed part of the linked.xml file. Full version of link.xml and description is here:

    You need to add a link.xml file under Assets/ folder. It should include the AWS Polly service. Note that packagenames should not include Experimental keyword.

    link.xml file:
    Code (CSharp):
    1. <linker>
    2. <!-- if you are using AWSConfigs.HttpClient.UnityWebRequest option-->
    3. <assembly fullname="UnityEngine">
    4.     <type fullname="UnityEngine.Networking.UnityWebRequest" preserve="all" />
    5.     <type fullname="UnityEngine.Networking.UploadHandlerRaw" preserve="all" />
    6.     <type fullname="UnityEngine.Networking.UploadHandler" preserve="all" />
    7.     <type fullname="UnityEngine.Networking.DownloadHandler" preserve="all" />
    8.     <type fullname="UnityEngine.Networking.DownloadHandlerBuffer" preserve="all" />
    9. </assembly>
    10. <assembly fullname="mscorlib">
    11.     <namespace fullname="System.Security.Cryptography" preserve="all"/>
    12. </assembly>
    13. <assembly fullname="System">
    14.     <namespace fullname="System.Security.Cryptography" preserve="all"/>
    15. </assembly>
    16. <assembly fullname="AWSSDK.Core" preserve="all"/>
    17. <assembly fullname="AWSSDK.CognitoIdentity" preserve="all"/>
    18. <assembly fullname="AWSSDK.SecurityToken" preserve="all"/>
    19. <assembly fullname="AWSSDK.Polly" preserve="all"/>
    20. </linker>
     
    Stefan-Laubenberger likes this.
  11. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Thank you!
     
  12. pabloHatzi

    pabloHatzi

    Joined:
    Jan 30, 2015
    Posts:
    7
    Hi, I try the example 4 with greek text. I Hear the sentence but the mark current spoked doesn't work.
    I modified the code like this :
    public void SpeakerA()
    { //Don't speak the text immediately
    uidSpeakerA = Speaker.Speak("Οι μελανιές, λοιπόν, που είχε αποκτήσει η Ελεονώρα Μελέτη ", SourceA, Speaker.VoiceForGender(Model.Enum.Gender.MALE, "el"), false, RateSpeakerA);
    }
    and get this error

    Could not speak the text: Wrapper {Uid='94fe7b69-a581-4848-861b-6ee765e7ca81', Text='Οι μελανιές, λοιπόν, που είχε αποκτήσει η Ελεονώρα Μελέτη ', Source='True', Voice='Vocalizer Expressive Nikos Harpo 22kHz (el-GR, MALE)', SpeakImmediately='True', Rate='1.1', Pitch='1', Volume='0', OutputFile='C:\Users\pablo\AppData\Local\Temp\DefaultCompany\LearnNumbersBraille\rtvoice_94fe7b69-a581-4848-861b-6ee765e7ca81.wav', ForceSSML='True', Created='3/21/2019 4:12:03 PM'}
    Exit code: 200
    Object reference not set to an instance of an object.
    at System.Speech.Internal.Synthesis.VoiceSynthesis.Speak(Prompt prompt)
    at Crosstales.RTVoice.TTSWrapper.TTSWrapper.speak(String speechText, SpeechSynthesizer speechSynthesizer)
    at Crosstales.RTVoice.TTSWrapper.TTSWrapper.speak(String[] args)
    at Crosstales.RTVoice.TTSWrapper.TTSWrapper.Main(String[] args)
    UnityEngine.Debug:LogError(Object)
    Crosstales.RTVoice.Provider.<SpeakNative>c__Iterator0:MoveNext() (at Assets/Plugins/crosstales/RTVoice/Scripts/Provider/VoiceProviderWindowsLegacy.cs:229)
    UnityEngine.SetupCoroutine:InvokeMoveNext(IEnumerator, IntPtr)
     
  13. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hi

    What happens if you do the same with a Windows built-in voice? I suspect something is wrong with the voice...

    Can you please try the text in our "01-Speech" demo scene with "Vocalizer Expressive Nikos Harpo 22kHz" and "Native" enabled/disabled?
    Does it work?


    So long,
    Stefan
     
  14. pabloHatzi

    pabloHatzi

    Joined:
    Jan 30, 2015
    Posts:
    7
    Hi Stefan
    I try with success your suggestion. In both cases (Native on, off) it works fine. I also saw that the event
    SpeakComplete _onSpeakComplete works fine but the event SpeakCurrentWord _onSpeakCurrentWord don't fire.
     
  15. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Have you also activated "Windows Legacy" on the RTVoice-component?
     
  16. pabloHatzi

    pabloHatzi

    Joined:
    Jan 30, 2015
    Posts:
    7
  17. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hello again

    I tried it with your voice and it seems to be the cause of the problem. I have original voices from Nuance, called "Vocalizer Expressive Allison Compact" and "Vocalizer Expressive Tom Compact" - after I installed your trial voice, they don't work anymore :(

    It also messes with the path variable and Unity gives me this error:
    upload_2019-3-21_23-53-32.png

    Unfortunately, we can't do anything against it, so please ask the producers of the voice.


    Cheers
    Stefan
     
  18. pabloHatzi

    pabloHatzi

    Joined:
    Jan 30, 2015
    Posts:
    7
    Hi Stefan

    To solve the problem that shows up after you install the voice from the link that I have send to you; you need to remove RT-Voice and then import it again. After this, you will be able to see the problem that occurs.
    Paul
     
  19. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hi Paul

    Hmm, I did encounter the same problem you described: Exit code 200
    Unfortunately, the Greek voices seems to break something in the SAPI-system.
    Please ask the provider of your voice for support.

    Have a nice weekend!


    Cheers
    Stefan
     
  20. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Awesome news:

    We managed to get the Windows native voice provider running under IL2CPP!
    Furthermore, eSpeak for Windows works now also with IL2CPP :)

    This features will be released with the next version 2019.2.1.
     
    Bartolomeus755 likes this.
  21. mindboxvr

    mindboxvr

    Joined:
    Mar 4, 2019
    Posts:
    1
    Hi all,
    I have one issue - how do you handle situation, when RT voice is speaking, and 3dparty app notification come, or phone call is incoming? I have Andoid 9 and Google Pixel and when incoming call - only small pop shows above my unity app. But when I hit dismiss, OnApplicationFocus or OnApplicationPause is not called. Same situation for notifications - only small popup appears and when I ignore this, it disapears.
    RTvoice stops speaking in these situations and because OnApplicationFocus or OnApplicationPause is not called, I dont know how to start speaking again - Speaker.speak.....

    Thanks.
     
  22. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hi

    I think I see the problem - you would like to Pause/UnPause on focus lost, isn't it?
    Please contact me via email so we can discuss this further.


    So long,
    Stefan
     
  23. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    We've just released version 2019.2.1 of RT-Voice.
    Changes:
    • Windows: Native and eSpeak voice providers are now working under IL2CPP!
    • New voice provider "SAPI Unity" added! See "3rd party"-folder for the integration
    • WSA: code improved and native mode added (when built as XAML without IL2CPP)
    • AWS Polly: "Auto Breath" added
    • Speaker: OnApplicationFocus now pauses and unpauses per default
    • Speaker and Provider: "isWorkingInPlaymode" and "isIL2CPPSupported" added
    • Voice: Version added
    • Editor improved
    • Updated to Common 2019.2.5
     
    Bartolomeus755 likes this.
  24. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    The new version is in the store!
     
    Bartolomeus755 likes this.
  25. roboroo

    roboroo

    Joined:
    Mar 27, 2017
    Posts:
    9
    Hi Stefan, installed the new update but now the RTvoice prefab comes up with an error "The current TTS only works in builds" Also the windows voices are missing. None of the scripts created with the original version will play the audio? Tried deleting all the original files and reinstalling but same error?
     
  26. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    What is you current build target?
     
  27. roboroo

    roboroo

    Joined:
    Mar 27, 2017
    Posts:
    9
    Android, using Unity 2018.3
     
  28. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    This is intentional - you are using Android, which has a totally different TTS-system than your development machine. This is why we decided to deactivate the TTS of the development machine and put the ""The current TTS only works in builds"-message (which is absoluetly true).
    If you switch to Windows, macOS or Linux as build target (depending on your os), you could test it again inside the editor.
    We did this because some users were confused about the different voices, compared between the builds and editor.

    However, I see you're making a point and imho the best option would be a flag in the configuration for the editor, like "Enforce Standalone TTS" (default: true).
    We will add this to the next release.


    So long,
    Stefan
     
    Last edited: Apr 19, 2019
  29. roboroo

    roboroo

    Joined:
    Mar 27, 2017
    Posts:
    9
    Ok, understand now, but it makes it hard to monitor the progress of the game build if you can't test the voices? I get it that the Android voices are in device dependent. The previous version worked fine for me, will have to rollback to that somehow..is that version available? Can email you if you have a link to it?. Thanks
     
  30. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    I've sent you an email with the previous version.
     
  31. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    We've just released version 2019.2.3 of RT-Voice.
    Main changes:
    • "Enforce Standalone TTS" for development in the configuration added (e.g. for mobile)
    • Fixes for iOS, Android and WSA build platforms
    • Better Editor handling
    • Tested with Unity 2019.1.0
    • Updated to Common 2019.3.0

    @roboroo this version contains the new flag "Enforce Standalone TTS"
     
    Bartolomeus755 likes this.
  32. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    The new version is in the store!
     
    Bartolomeus755 likes this.
  33. kilik128

    kilik128

    Joined:
    Jul 15, 2013
    Posts:
    843
    hum how espeak ng must be use ?
     
  34. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hi

    You asked about Klattersynth before. "Speech.cs" is located here:
    upload_2019-4-25_23-58-49.png

    Anyway, to use eSpeak-NG, you have to edit "Constants.cs" and change the variable "DEFAULT_TTS_LINUX" (starting at line 113):
    Code (CSharp):
    1. #if !ENABLE_IL2CPP && (UNITY_STANDALONE_WIN || UNITY_EDITOR_WIN)
    2.         //public const string DEFAULT_TTS_LINUX = "C:\\Program Files (x86)\\eSpeak\\command_line\\espeak.exe"; // eSpeak under Windows
    3.         public const string DEFAULT_TTS_LINUX = "C:\\Program Files\\eSpeak NG\\espeak-ng.exe"; // eSpeak-NG under Windows
    4. #else
    5.         //public const string DEFAULT_TTS_LINUX = "espeak";
    6.         public const string DEFAULT_TTS_LINUX = "espeak-ng";
    7. #endif
    8.  
    9.         public const string DEFAULT_TTS_LINUX_DATA = "";
    10.         //public const string DEFAULT_TTS_LINUX_DATA = @"C:\Program Files (x86)\eSpeak\espeak-data\"; //  eSpeak data under Windows
    11.         //public const string DEFAULT_TTS_LINUX_DATA = @"C:\Program Files\eSpeak NG\espeak-ng-data"; // eSpeak-NG data under Windows
    I'm not 100% sure if you also have to change "DEFAULT_TTS_LINUX_DATA" to @"C:\Program Files\eSpeak NG\espeak-ng-data"

    Please let me know if it works for you.


    So long,
    Stefan
     
  35. kilik128

    kilik128

    Joined:
    Jul 15, 2013
    Posts:
    843
    got this result
    this so i'am not sure
    what's must do after





    devenv_2019-04-26_09-52-34.jpg Unity_2019-04-26_09-52-10.jpg
     
  36. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    You have currently "MaryTTS" enabled. Please set it like this:
    upload_2019-4-26_11-7-40.png

    eSpeak is active if you have more than 80 voices.
     
  37. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    We've just released version 2019.3.0 of RT-Voice.
    Main changes:
    • New voice provider WebGL added! See "3rd party"-folder for the integration
    • Windows: voice selection fixed
    • Pattern matching for cultures simplified. E.g. "en US", "enus", "en-us" and "en_us" is now working.
    • PlayMaker actions "GetVoices" and "GetCultures" added
    • Updated to Common 2019.3.1
     
    Bartolomeus755 likes this.
  38. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    The new version is in the store!
     
  39. rgonsalv

    rgonsalv

    Joined:
    Sep 29, 2016
    Posts:
    37
    We were hoping to be able to use Acapela voices on Android. Even though their engine is available in Settings, the voices don't appear in Speaker.Voices list. Is this something that could be implemented? One special reason to support this is that they support Voice Banking, where a user can create a voice based on recording some sample text.
     
  40. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hi

    RT-Voice uses the default TTS-classes from Android:
    https://developer.android.com/reference/android/speech/tts/TextToSpeech

    If voices don't show up, it's a problem with the engine...
    We had those issues before with non-Google engines and unfortunately it looks like Acapela is also not properly supported. :(
    This is beyond our reach and we can't really help you further. Probably this could help (or Acapela itself):
    https://www.acapela-group.com/acapela-tts-voices-googleplay-instructions/


    Cheers
    Stefan
     
  41. rgonsalv

    rgonsalv

    Joined:
    Sep 29, 2016
    Posts:
    37
    It turns out that I was wrong! I can get the Acapela voices to show up and they will work with your API.
     
    Stefan-Laubenberger likes this.
  42. kmk68

    kmk68

    Joined:
    Feb 14, 2014
    Posts:
    38
    Hello!

    I bought RT-Voice recently and it's great! Or rather - he was ...I installed it in the project - I am going to use it mainly to read the helps (manuals) in the game, give advice, etc. He installed ok, I left him installed in the project, reading only a short greeting. To know if any next asset will cause a conflict. But it was ok.

    Three days ago, a computer crasch (or Unity) occurred - just during the playback of the audio message. The program released the sound "aggggggghhhrrr" ... and the end. The Unity project has disappeared. The tree of the project was no longer visible ... etc.

    Everything returned to normal work when I deleted RT-Voice from the project. And now I can not install it again - it ends with errors like in the picture below.

    Interestingly - I do not know how the assets are created - but when I return to the version of the earlier project (I have everyday backups) also the attempt to install RT-Voice ends with an error as below.

    Regardless of the platform set up. :-( Maybe you can help me something?

    Windows 10, Unity 5.6.7f1. I operate on iOS / Android / WSA - mobile platforms only.

    Regards,
    Kajetan


    upload_2019-5-24_19-0-48.png
     
  43. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hi Kajetan

    It seems your Unity-crash destroyed our preferences file. :(

    Can you please try this:
    1. go to Tools -> RT-Voice -> Configuration
    2. a window opens, go to the tab "Config"
    3. press the "Save"-button at the bottom
    4. Restart Unity

    Please let me know if it helps.


    So long,
    Stefan
     
  44. kmk68

    kmk68

    Joined:
    Feb 14, 2014
    Posts:
    38
    Sorry, but how to open the "config" after installing RT-Voice, the config screen is empty. :-( (after crasch of course).
    Maybe something must be removed? Some configuration file? Maybe bad (not entirely) uninstalling?

    Best regards,
    Kajetan
     
  45. kmk68

    kmk68

    Joined:
    Feb 14, 2014
    Posts:
    38
    Hello.

    I try just all: reimport RT-Voice, reinstall all asset, reinstall Unity...
    In new empty project is install ok. In this project - see below.

    My project before install RT-Voice
    upload_2019-5-25_21-33-1.png

    After install
    upload_2019-5-25_21-34-7.png

    Config window
    upload_2019-5-25_21-34-47.png

    And how I uninstall it:
    upload_2019-5-25_21-35-25.png

    Kind regards,
    Kajetan
     
  46. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Hello again

    Please delete the file "crosstales.cfg" under the following path (the same as "Application.persistentDataPath"):
    %USERPROFILE%/AppData/LocalLow/<company>/<product>/crosstales.cfg

    Please let me know if it helps.


    So long,
    Setfan
     
  47. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    We've just released version 2019.3.1 of RT-Voice.
    Main changes:
    • Linux: problems under non-Windows platforms solved
    • Rare problems with loading a corrupted configuration fixed
    • Updated to Common 2019.3.2

    @kmk68 this version also fixes the problems with the corrupted configuration
     
  48. kmk68

    kmk68

    Joined:
    Feb 14, 2014
    Posts:
    38
    Yes! Thanks a lot! :) Deleting the crosstales.cfg file has helped.

    Version 2019 ... great, but I hope that it will continue to work under 5.6
    I have a big project which I finish, which I can not bring back for 2019.

    RT-Voice is amazing!

    Thank you very much for your help.

    Best regards,
    Kajetan
     
    Stefan-Laubenberger likes this.
  49. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    Thank you for the flowers :)
    If you like RT-Voice, please don't forget to rate it or write a review - it's very much appreciated!

    I also take the opportunity to advertise our tool Turbo Backup - if you're unsure if your project will work seamless in 2019, make a backup and if everything fails, just go back to 5.6.
     
  50. Stefan-Laubenberger

    Stefan-Laubenberger

    Joined:
    May 25, 2014
    Posts:
    1,296
    The new version is in the store!