createSttModel abstract method
Future<SpeechRecognizer>
createSttModel({
- String? modelPath,
- String? tokenizerPath,
- PreferredBackend? preferredBackend,
Creates and returns a new SpeechRecognizer instance.
Modern API: If paths are not provided, uses the active STT model set via
FlutterGemma.installStt() or modelManager.setActiveModel().
modelPath — path to the STT model file (optional if active model set).
tokenizerPath — path to the tokenizer file (optional if active model set).
preferredBackend — backend preference (e.g., CPU, GPU).
Implementation
Future<SpeechRecognizer> createSttModel({
String? modelPath,
String? tokenizerPath,
PreferredBackend? preferredBackend,
});