How to use Speech to Text to Speech
IMPORTANT NOTE: No matter which speech recognition strategy you use: MAKE SURE THAT THE SPEECH RECOGNITION INDICATOR IN THE TOP LEFT CORNER OF YOUR SCREEN IS NOT SHOWING ALL THE TIME!!! This means you have configured your settings incorrectly and can lead to high billing costs. I made sure this never happens with the default values but follow this guide CAREFULLY just in case.
This guide assumes you already created and imported Google Cloud service account credentials. Follow this guide if that's not the case.
Once Google Cloud account credentials are imported, you can enable the feature in the settings under "Audio Input".
There are 2 speech recognition strategies you can use for Speech-to-Text-to-Speech:
- Push-to-record
- Automatic speech detection
- Select the language you want your speech to be recognized in
- Select the input device you want to use as "Recording device"
- Select "Push-to-record" as "Speech recognition stragegy"
- Set a key to press in order to record your speech
You can now press the registered key, speak into the selected recording device then release the registered key to transcribe speech. There should be an icon in the top left corner of your screen indicating you that your speech is being recorded.
- Select the language you want your speech to be recognized in
- Select the input device you want to use as "Recording device"
- Select "Automatic" as "Speech recognition stragegy"
- Set the amount of volume required in order to activate speech recording (in dB). Make sure the volume of ambient sound around you is BELOW that value and the volume of your speech ABOVE it. Make sure the speech recognition indicator is NOT showing constantly in the top left corner of your screen.
You can now speak into the selected recording device to transcribe speech. There should be an icon in the top left corner of your screen indicating you that your speech is being recorded.