Metahuman Convai Character not responding text box when I push down T Key

Unreal Engine Questions
1

I‘m having an issue getting a response from the Convai character. I’m using UE5.6. I have a Metahuman character. The character responds to text input but not voice conversation. Voice chat is not showing in the text box. I have troubleshooted the microphone, and everything is working. I tested on Convai Playground, and voice input works there. I receive a string output when I push down the T key on the screen, and the output log. Not sure how to fix the issue. Can you help with this matter? Thaks!

Output log:
ConvaiSubsystemLog: Start Run
ConvaiSubsystemLog: UConvaiSubsystem Started
LogChaosDD: Creating Chaos Debug Draw Scene for world DefaultLevel
LogPlayLevel: PIE: World Init took: (0.002628s)
LogAudio: Display: Creating Audio Device: Id: 14, Scope: Unique, Realtime: True
LogAudioMixer: Display: Audio Mixer Platform Settings:
LogAudioMixer: Display: Sample Rate: 48000
LogAudioMixer: Display: Callback Buffer Frame Size Requested: 1024
LogAudioMixer: Display: Callback Buffer Frame Size To Use: 1024
LogAudioMixer: Display: Number of buffers to queue: 1
LogAudioMixer: Display: Max Channels (voices): 32
LogAudioMixer: Display: Number of Async Source Workers: 4
LogAudio: Display: AudioDevice MaxSources: 32
LogAudio: Display: Audio Spatialization Plugin: None (built-in).
LogAudio: Display: Audio Reverb Plugin: None (built-in).
LogAudio: Display: Audio Occlusion Plugin: None (built-in).
LogAudioMixer: Display: Initializing audio mixer using platform API: ‘XAudio2’
LogAudioMixer: Display: Using Audio Hardware Device Headset Earphone (Plantronics Blackwire 5220 Series)
LogAudioMixer: Display: Initializing Sound Submixes…
LogAudioMixer: Display: Creating Master Submix ‘MasterSubmixDefault’
LogAudioMixer: Display: Creating Master Submix ‘MasterReverbSubmixDefault’
LogAudioMixer: FMixerPlatformXAudio2::StartAudioStream() called. InstanceID=14
LogAudioMixer: Display: Output buffers initialized: Frames=1024, Channels=2, Samples=2048, InstanceID=14
LogAudioMixer: Display: Starting AudioMixerPlatformInterface::RunInternal(), InstanceID=14
LogAudioMixer: Display: FMixerPlatformXAudio2::SubmitBuffer() called for the first time. InstanceID=14
LogInit: FAudioDevice initialized with ID 14.
LogAudio: Display: Audio Device (ID: 14) registered with world ‘DefaultLevel’.
LogAudioMixer: Initializing Audio Bus Subsystem for audio device with ID 14
LogLoad: Game class is ‘GM_Sandbox_C’
LogWorld: Bringing World /Game/Levels/UEDPIE_0_DefaultLevel.DefaultLevel up for play (max tick rate 60) at 2025.10.29-21.46.25
LogWorld: Bringing up level for play took: 0.050225
LogOnline: OSS: Created online subsystem instance for: :Context_27
LogVoiceEncode: Display: EncoderVersion: libopus unknown
LogOnlineVoice: OSS: StopLocalVoiceProcessing(0) returned 0xFFFFFFFF
LogOnlineVoice: OSS: Stopping networked voice for user: 0
ConvaiPlayerLog: UConvaiPlayerComponent: Found submix “AudioInput”
LogSkeletalMesh: USkeletalMeshComponent: Recreating Clothing Actors for ‘TwinBlast’ with ‘KOM2_UECC4’
Cmd: voice.MicNoiseGateThreshold 0.01
voice.MicNoiseGateThreshold = “0.01”
Cmd: voice.SilenceDetectionThreshold 0.001
voice.SilenceDetectionThreshold = “0.001”
PIE: Server logged in
PIE: Play in editor total start time 0.312 seconds.
LogBlueprintUserMessages: [MicSettings_WB_C_0] Device Set Succesfully
LogBlueprintUserMessages: [CBP_SandboxCharacter_KOM2CC4_C_0] Hello
LogBlueprintUserMessages: [CBP_SandboxCharacter_KOM2CC4_C_0] Hello
LogBlueprintUserMessages: [CBP_SandboxCharacter_KOM2CC4_C_0] Hello
LogBlueprintUserMessages: [CBP_SandboxCharacter_KOM2CC4_C_0] Hello
LogSlate: Updating window title bar state: overlay mode, drag disabled, window buttons hidden, title bar hidden
LogWorld: BeginTearingDown for /Game/Levels/UEDPIE_0_DefaultLevel
LogCrowdFollowing: Warning: Unable to find RecastNavMesh instance while trying to create UCrowdManager instance
ConvaiSubsystemLog: UConvaiSubsystem Stopped
LogWorld: UWorld::CleanupWorld for DefaultLevel, bSessionEnded=true, bCleanupResources=true
ConvaiSubsystemLog: End Run
LogSlate: InvalidateAllWidgets triggered. All widgets were invalidated
LogPlayLevel: Display: Shutting down PIE online subsystems
LogSlate: InvalidateAllWidgets triggered. All widgets were invalidated
LogSlate: Updating window title bar state: overlay mode, drag disabled, window buttons hidden, title bar hidden
ConvaiChatbotComponentLog: Cleanup | Character ID : ad7ae05c-9e64-11f0-abc0-42010a7be025 | Session ID : -1
LogAudioMixer: Deinitializing Audio Bus Subsystem for audio device with ID 14
LogAudioMixer: Display: FMixerPlatformXAudio2::StopAudioStream() called. InstanceID=14, StreamState=4
LogAudioMixer: Display: FMixerPlatformXAudio2::StopAudioStream() called. InstanceID=14, StreamState=2
LogUObjectHash: Compacting FUObjectHashTables data took 2.17ms
LogPlayLevel: Display: Destroying online subsystem :Context_27
LogEOSSDK: LogEOS: Updating Product SDK Config, Time: 6237.169922
LogEOSSDK: LogEOS: SDK Config Product Update Request Completed - No Change
LogEOSSDK: LogEOS: ScheduleNextSDKConfigDataUpdate - Time: 6237.262695, Update Interval: 356.608795

image
image1142×1030 131 KB

image
image797×625 97.8 KB

image
image1125×455 117 KB

2

Hello,

Welcome to the Convai Developer Forum!

So just to clarify, are you saying that speech-to-text (STT) isn’t working? You’re speaking, but the chat box doesn’t display your spoken text, correct?

Also, when you test your microphone in the Settings menu, does it register input properly?

3

Hello,
Yes, STT is not working correctly. When I start the engine, it displays “Device set successfully “. When I go to settings to test. I click the record, but don’t hear anything back to let me know it’s working. Is there any other information I can send you to help troubleshoot this matter?
Thanks!

4

This indicates that your microphone isn’t working properly within Unreal Engine. Please make sure you’ve selected the correct microphone device and that Unreal Engine has the necessary microphone permissions on your system.

This issue isn’t related to Convai’s STT functionality. We recommend checking your microphone settings or reviewing relevant discussions on the Unreal Engine forums. You can also use tools like ChatGPT for additional troubleshooting guidance.

5

So I guess I’m not understanding the issue because when I add an (Audio Capture Component) to the player and run the game, I hear my voice audio back in the engine when I talk. Not sure why it’s not recognizing my voice in the text box when communicating with the Convai character.