OMG thank you for the fast reply! Really appreciate what are you doing!!!
Documentation updated: Wiki (github.com)
Hi, lucoiso. Thank you very much for your efforts, but in version 1.6.8, I found that the two blueprint nodes Speech to Text with Custom Options and Speech to Text with Default Options have not changed, and the problem is still the same. I can’t control the input of the microphone. So how to enable the new function “Disconnect signals before trying to end the task to avoid undesired updates while closing the connection” you mentioned?
LogAzSpeech_Internal: Error: Thread: AzSpeech_SpeechToText_177663; Function: ProcessRecognitionResult; Message: Task failed. Reason: NoMatch
My mic is a usb Bluetooth mic, It’s Good in Windows system.
Got this error in azspeech
If routed through voicemeter, it is good.
Can you help me fix it?
Hello, Lucoiso!
The plugin work great! But is any way to know which locale is speaking? I am using 2 locales: en-US, vn-VN and using language identification.
It is like a runtime recognizes voice or a runtime recognizes local. I saw that you have a runtime that recognizes strings in speech-to-text but not in text-to-speech.
Hope you will make that function in your next update!
@lucoiso I find this log :
Audio Buffer Underrun (starvation) detected.
My mic is a usb Bluetooth mic, It’s Good in Windows system.
Got this error in azspeech
If routed through voicemeter, it is good.
Can you help me fix it?
Hello there!
Has anyone used the “TextToWavFile”?
I sometimes get corrupted audio files when I use it.
btw. lucoiso you’re doing an amazing job. We are using your Chat Gpt plugin as well and it works wonders with this one. 
Hi Im trying to just start to making it work but it always fail, can someone tell me what is wrong with my blueprint?
For BP2 image, maybe you can refer to the sample project, and leave the text areas of Device ID and List Group as blank. And if you want to enter the language you want in Locale, please refer to your project setting, and copy & paste one language you prefer (eg. en-US) to the text area. Hope this will help~
Hello, I wouId like to try your plugin but I am new to the UE world and I am unable to make it works.
I have tried to open the sample projects but none of them open on my side (W11, UE5.1.1, VS2022+GamingModule). I am asked to rebuild the modules when I open each projects, so I click yes but it fails and tell me to rebuild from source for both UESpeechGPT and UEAzSpeechSampleProject.
I have also tried to use directly the plugin functions with the tiny examples here and there on this thread without success, here is my basic blueprint FYI.
The logs does not help me to figure out where the problem is coming from. It tells it starts but nothing more, there is no warning or feedback displayed after the “Starting Recognition” message.
I have activated all logs if it helps someone:
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145241; Function: Run; Message: Recognition started.
LogAzSpeech: Display: Task: SpeechToText (145532); Function: Activate; Message: Activating task
LogAzSpeech_Internal: Display: Task: SpeechToText (145532); Function: StartAzureTaskWork; Message: Starting Azure SDK task
LogAzSpeech_Internal: Display: Task: SpeechToText (145532); Function: StartAzureTaskWork; Message: Using audio input device: Default
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: Init; Message: Initializing runnable thread
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: CanInitializeTask; Message: Checking if can initialize task in current context
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: Run; Message: Running runnable thread work
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: InitializeAzureObject; Message: Initializing Azure Object
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: InitializeAzureObject; Message: Creating recognizer object
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: CreateSpeechConfig; Message: Creating Azure SDK speech config
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: ApplySDKSettings; Message: Applying Azure SDK Settings
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: EnableLogInConfiguration; Message: Enabling Azure SDK log
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: InsertProfanityFilterProperty; Message: Adding profanity filter property
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: ApplySDKSettings; Message: Using language: Default
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: Run; Message: Starting recognition
LogBlueprintUserMessages: [NewMap_C_9] Started !!!!!!!!!!!!!!!!
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: Run; Message: Recognition started.
LogAzSpeech: Display: Task: SpeechToText (145532); Function: PrePIEEnded; Message: Trying to finish task due to PIE end
LogAzSpeech: Display: Task: SpeechToText (145532); Function: StopAzSpeechTask; Message: Stopping task
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: StopAzSpeechRunnableTask; Message: Setting runnable work as pending stop
LogAzSpeech: Display: Task: SpeechToText (145532); Function: SetReadyToDestroy; Message: Setting task as Ready to Destroy
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: StopAzSpeechRunnableTask; Message: Setting runnable work as pending stop
LogAzSpeech: Display: Task: SpeechToText (145241); Function: PrePIEEnded; Message: Trying to finish task due to PIE end
LogAzSpeech: Display: Task: SpeechToText (145241); Function: StopAzSpeechTask; Message: Stopping task
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145241; Function: StopAzSpeechRunnableTask; Message: Setting runnable work as pending stop
LogAzSpeech: Display: Task: SpeechToText (145241); Function: SetReadyToDestroy; Message: Setting task as Ready to Destroy
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145241; Function: StopAzSpeechRunnableTask; Message: Setting runnable work as pending stop
LogSlate: Updating window title bar state: overlay mode, drag disabled, window buttons hidden, title bar hidden
LogWorld: BeginTearingDown for /Game/UEDPIE_0_NewMap
LogWorld: UWorld::CleanupWorld for NewMap, bSessionEnded=true, bCleanupResources=true
LogSlate: InvalidateAllWidgets triggered. All widgets were invalidated
LogContentBundle: [Container: NewMap] Deleting container.
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: Stop; Message: Stopping runnable thread work
LogPlayLevel: Display: Shutting down PIE online subsystems
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145241; Function: Stop; Message: Stopping runnable thread work
LogSlate: InvalidateAllWidgets triggered. All widgets were invalidated
LogSlate: Updating window title bar state: overlay mode, drag disabled, window buttons hidden, title bar hidden
LogAudioMixer: FMixerPlatformXAudio2::StopAudioStream() called. InstanceID=17
LogAudioMixer: FMixerPlatformXAudio2::StopAudioStream() called. InstanceID=17
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145241; Function: Stop; Message: Stopping runnable thread work
LogAzSpeech_Internal: Display: Thread: AzSpeech_SpeechToText_145532; Function: Stop; Message: Stopping runnable thread work
LogUObjectHash: Compacting FUObjectHashTables data took 1.09ms
LogPlayLevel: Display: Destroying online subsystem :Context_18This text will be hidden
I am a bit lost here, do you have any idea that could help me ?
I finally managed to make it works; I used the git release version 1.6.8 instead of the marketplace one and was able to shift to UE5.2 at the same time. yeah !!
Great plugin by the way @lucoiso
Hey! I dont know why, but every function that I use never start. Im trying the “.wav file to text custom” and nothing happens, but in the output log I have this:
I have setted the Settings (Subscription Key and Region ID).
Thanks!
Goodmorning!
I want to ask if this plugin support the “streaming speech recognition” that azure allow.
May I ask why my Speech to Text with Default Options node is not working and the project settings have been configured
looks like the audio input device is invalid, the correct format is something like: {1234.5678…}
after a long time without a computer since i moved from my country, a new update xd:
→ Demo Updated: Release AzSpeechSampleProject v1.0.0 · lucoiso/UEAzSpeechSampleProject · GitHub
Note: I can only update the product on the marketplace to the latest 3 Engine versions, so UE4.27 is only being updated on GitHub!
Thank you very much for your plugin which is really useful and strong. Recently, i va been trying to figure out the pin called async task. i am wondering what is the usage of that. And how can we use it? looking foward seing your words.
