Speach Recognition could not start error on Remote Desktop Server

Question

When we log into our remote desktop server as a normal user we get the message that: Speach Recognition could not start becuase the language configuration is not supported.

When we log in as an Administrator we do not see this message and it works fine.

Answer 1

Hello Bryan Colby1,

I understand that you could not start Speach Recognition as a normal user.

This problem is usually because the normal user's profile lack the necessary language and speech recognition settings, or the normal user account does not have permission to access or use system services or resources.

You can solve the problem by following these steps:

1. Group Policy Review:

    Open the Group Policy Editor (gpedit.msc).

    Investigate the following policies that may affect speech recognition:

      Computer Configuration > Administrative Templates > Windows Components > Speech Recognition

      User Configuration > Administrative Templates > Control Panel > Regional and Language Options

      User Configuration > Administrative Templates > System > Internet Communication Management > Internet Communication settings

      Computer Configuration > Policies > Administrative Templates > System > Services

2. File and Registry Permissions:

    Check the permissions of files and subfolders within the C:\Windows\System32\speech directory, ensuring that nonadministrative users have at least read and execute permissions.

3. Check Language and Speech Recognition Settings:

    Log in to the RDS server using the affected nonadministrator account (or simulate the session).

    Open "Control Panel" > "Time & Language" > "Region & Language" > "Administrative Language Settings" and ensure the supported speech recognition language is installed and set as the preferred language.

    Check the "TexttoSpeech" tab under "Language options" to confirm if the required language's speech synthesis engine is installed and enabled

4. Examine the Speech Recognition Service:

    Launch the "Services" management console (services.msc).

    Locate the "Windows Speech Recognition (WSService)" service, make sure its Startup type is set to "Automatic," and attempt to start the service.

    Verify that the service login account has sufficient permissions to run the service; if not, change it to an account with adequate permissions (such as the Local System account).

If necessary, inspect registry key permissions related to speech recognition to ensure the user account has sufficient access rights.

I hope the information above is helpful. If you have any question or concern, please feel free to let us know.

Best Regards,

Xin Wang

Answer 2

It was helpful. We went through and made some changes but we are still having the same issue.

There were some items in your list that are a little different on windows 2022 then what you had listed.

We still get an error that says:

Answer 3

Xin Wang, here are our verification and troubleshooting results. We walked through the recommendations you sent:

1.    Group Policy Review (Open the Group Policy Editor (gpedit.msc). Investigate the following policies that may affect speech recognition)

1. Computer Configuration > Administrative Templates > Windows Components > Speech Recognition – Verified options are either enabled or not configured to limit
2. User Configuration > Administrative Templates > Control Panel > Regional and Language Options – Verified options are either enabled or not configured to limit
3. User Configuration > Administrative Templates > System > Internet Communication Management > Internet Communication settings – Verified options are either enabled or not configured to limit
4. Computer Configuration > Policies > Administrative Templates > System > Services – Could not locate this path in group policies for local or domain
   1. File and Registry Permissions
      1. Check the permissions of files and subfolders within the C:\Windows\System32\speech directory, ensuring that nonadministrative users have at least read and execute permissions. – Permissions verified
   2. Check Language and Speech Recognition Settings (Log in to the RDS server using the affected nonadministrator account (or simulate the session))
      1. Open "Control Panel" > "Time & Language" > "Region & Language" > "Administrative Language Settings" and ensure the supported speech recognition language is installed and set as the preferred language.
      2. Time & Language don’t exist in Control Panel, verified following:
         1. Region, language preferences, Windows Display language “English (United States)
         2. Region, Language preferences, Preferred Language “English (United States)”
         3. Region, Country or region “United States”
         4. Region, Regional format “English (United States)”
      3. Check the "TexttoSpeech" tab under "Language options" to confirm if the required language's speech synthesis engine is installed and enabled
         1. Speech, Speech language “English (United States)”
         2. Speech, Installed voice packages “English (United States)” and “English (Canada)”
   3. Examine the Speech Recognition Service (Launch the "Services" management console (services.msc). Locate the "Windows Speech Recognition (WSService)" service, make sure its Startup type is set to "Automatic," and attempt to start the service.
      1. Windows Speech Recognition Service not installed
      2. Windows Audio service set to automatic and running
      3. Windows Audio Endpoint Builder service set to manual and running

We are still receiving the error message Bryan sent previously. Please review our results and let us know what you recommend.

Answer 4

Hello,

Sorry for the late reply.

Since none of the previous methods worked, you can continue to troubleshoot as follows：

Sometimes, speech recognition components may become damaged or missing. You can try re-registering these components:

• Open Command Prompt (run as administrator).
• Type the following commands and press Enter after each one:

regsvr32.exe /u sspiclient.dll

regsvr32.exe sspiclient.dll

Besides, Check registry entries.

Certain registry entries may affect speech recognition functionality. You can try modifying or deleting these entries:

Please note that directly editing the registry can lead to system issues, so proceed with caution. If you're unsure how to proceed, seek assistance from a professional.

Here are the steps to delete or rename the registry keys:

• Open Registry Editor, and navigate to the registry key:

HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Speech

• Delete the registry keys:

In the right-hand pane, right-click on "UserProfiles" and "Voices" keys and select "Delete."

Close the Registry Editor and restart your computer. Windows will automatically recreate these registry keys upon restarting the system.

Please let me know if the issues are still not resolved.

Regards,

Xin Wang