Here is the demo from Apple's site
This issues is specific to iOS 18.
When running this demo, we are getting new text when we have a gap in speaking, the recognitionTask(with:resultHandler:) provides new text which is only spoken after the gap and not the concatenation of old text and the new spoken text.
Hello @Nishchal, thank you for your post. This issue was addressed in iOS 18.1 Beta 5 (Build: 22B5054e). Please test on the latest betas and, if the issue persists, please use Feedback Assistant to submit a bug report.
Hello @Engineer , I am facing the same issue on 17.5.1 when i use, SFSpeechRecognizer.supportsOnDeviceRecognition = true SFSpeechAudioBufferRecognitionRequest.requiresOnDeviceRecognition = true.