UE4 Google Speech Recognition Plugin including Voice Capture Function.
Engine version: 4.25.4
Platform: Windows 10
- Follow the instruction to set up a Cloud Console project and enable the Cloud Speech-to-Text API for that project.
- Create an API key
Create an Api key
- Set Api Key
- Add Recording Component to your actor
- Record your speech and test your microphone
- Set recognition configuration and send your speech to google
- Enable exception for packaged builds
Go to your {ProjectName}.Target.cs and add this in the constructor:
bForceEnableExceptions = true;
bOverrideBuildEnvironment = true;
- Speak no more than one minute each time.
- Capture audio with a sampling rate of 16,000 Hz or higher. And smaller sample rate will help you get the result faster (In general, 16,000 Hz is enoough).
- The SampleRate in GoogleSTTconfig should be identical with the Recording.
- You can add words, phrases or class tokens to Speech Contexts to help Speech-to-Text recognize specific words or phrases more frequently than other options that might otherwise be suggested.
- See more about config in documentation.