I would like to use both an Image Classifier and a Sound Classifier in my application. Since I have a video stream in use for both image analysis and speech recognition, I'd like to use the same AVCapture related APIs to analyze sound. I don't seem able to do that since the analyze function of the stream analyzer requires the audio format and frame position. I've only seen SNAudioStreamAnalyzer examples with the AVAudioEngine and node.installTap APIs. Is it possible to use AVCaptureSession and public func captureOutput(_ output: AVCaptureOutput, didOutput sampleBuffer: CMSampleBuffer, from connection: AVCaptureConnection) to trigger SNAudioStreamAnalyzer.analyze?