Google Cloud Speech-to-Text adds a cloud transcription engine backed by Google’s speech recognition models. The plugin authenticates with a Google Cloud service-account JSON key instead of a simple API key and stores credentials securely in the macOS Keychain. Long recordings are automatically split into overlapping chunks so they stay within Google’s per-request size limit.
Features
30+ languages including Arabic, Bengali, German, English, Spanish, French, Hindi, Japanese, Korean, Portuguese, Russian, Chinese, and more
Automatic punctuation and capitalization
Word-level time offsets for precise alignment
Speech context boosting for domain-specific vocabulary
Automatic audio chunking with overlap for long recordings
Built-in credential validation against the live API
Secure service-account storage in the macOS Keychain
Configurable default language code
Transcription Models
Model
ID
Default
default
Command and Search
command_and_search
Latest Short
latest_short
Latest Long
latest_long
Phone Call
phone_call
Video
video
Configuration
Service Account JSON - Create a service account in Google Cloud Console, generate a JSON key, and paste the full contents into the plugin settings. Simple API keys are not supported.
Model - Select a recognition model. Use default or command_and_search for the broadest language coverage.
Default Language - A BCP-47 language code (e.g. en-US, de-DE) used when TypeWhisper does not pass a spoken language.
Setup
Enable the Cloud Speech-to-Text API in your Google Cloud project
Create a service account and download the JSON key file
Open TypeWhisper Settings > Plugins
Find the Google Cloud Speech-to-Text plugin and click Configure
Paste the full JSON key contents and save
Select Google Cloud Speech-to-Text as your transcription engine in Settings or in a profile
