iOS 27 Siri is using WaveRNN and FastSpeech2 [D]
Siri on iOS 27 is reportedly utilizing WaveRNN and FastSpeech2 models, discovered within the simulator's files. These models are integrated in an espresso format, suggesting an optimized deployment for on-device processing. Additionally, a CoreML model for concert ranking, identified as a logistic regression, was also found. AI
IMPACT Enhances Siri's voice synthesis and potentially other audio-related features on iOS devices.