With rumors circulating about Apple looking deeper into voice as a progression of the iOS user interface on the iPhone and a possible Nuance acquisition, we’re beginning to see how voice may shape the future of mobile computing. A recent patent granted to Apple demonstrates how the company intends to use text to speech and speech to text to enhance the phone call capabilities of the iPhone.
In one scenario in a loud environment, users may have to shout into their phones to communicate with the other party. A more intelligent solution proposed by Apple would be to use text to speech where you would type what you would want to say, and the iPhone would then speak the text to the other party using TTS technology.
This service could be activated automatically if the iPhone senses the surrounding noise level to be above a certain decibel level and prompt the user to use TTS technology, or it could play back to the other party a pre-recorded message saying that due to the surrounding noise levels, the call cannot be made.
In another instance where users may be in a situation such as a meeting where they cannot take a call or speak into their phones. In this instance, speech to text technology would allow the other party to speak what they are trying to say and have the receiving party–who’s stuck in a meeting–receive the message as a text message on their iPhone. This transcription service would allow the user then, to reply, via TTS technology so they can maintain communication in a semi-natural way given the constraints of the environment.
And unlike similar TTS systems from providers like Nuance, it seems that all the transcription and processing work would be done on the device natively rather than sent to a server. Both the TTS and speech to text transcription engines would run off of the device’s application processor, like the Apple A4 and A5 chipsets.
You can read more about the patent through Patently Apple.