Speech synthesis

[youtube=http://www.youtube.com/watch?v=UOCA-nqIzz4&feature=player_embedded]

Since the introduction of Nuance speech technology in OS X Lion, which provides several new voice options in system preferences for the System Voice, many have compared the voice of Nuance-powered Siri on iPhone 4S to the improved text-to-speech included as free downloads within Lion. As noted by Reddit user Moosehadley, what you might not have realized is the downloadable “Samantha” voice for Lion is the same as Siri’s in the United States. Here is how to download it:

Open System Preferences> Speech> Text to Speech>System Voice> Customize> and select “Samantha” from the list. Apple will ask you to confirm the 469MB install.

Expand
Expanding
Close

Update: Some are calling this a fake, which technically wouldn’t be hard to do.

[youtube=http://www.youtube.com/watch?feature=player_embedded&v=xFIRmnRHNUM]

Apart from a press of the home button to initiate Siri, this prototype Siri hack first spotted by IntoMobile is completely thought-controlled. The guys behind Project Black Mirror have recorded brain wave activity with ECG pads, matched the incoming patterns to pre-saved digital patterns saved on a MacBook, then fed the matched commands to a speech synthesizer chip that translates the command to Siri. The video above shows the developers initiating a call, but they say they’ve linked approximately twenty-five brain wave patterns to various Siri-controlled functions, and hope to bypass having to physically press the home button with a fully automated solution in the future. IntoMobile breaks down the intricacies:

Expand
Expanding
Close