Apple’s SVP of worldwide market Phil Schiller sat down for a brief interview with Sound & Vision talking about the HomePod, why Apple decided to create it and why it will stand out from the competition.
For the very first question “Why HomePod?”, Phil says that it brings together all the work Apple has been doing to deliver an outstanding music experience in one’s home. Apple’s audio engineers and scientists have been pushing the boundary in term of what’s capable with the quad-speaker system on the iPad Pro and the AirPods. On the other hand, Apple Music has turned into a great music streaming service and Siri has become smart enough to help you discover new music based on your taste. Add in the technological advancements and machine learning, and you get the HomePod — a device where all these technologies and advancements work together to deliver fantastic music experience.
Schiller then talks about various HomePod features and how Apple is able to create spatial awareness to deliver great music quality despite the tiny form factor of the device.
From the moment you plug in HomePod and start listening to music, it automatically and instantly senses its location in the room to take full advantage of the environment it’s in. The microphone array in HomePod listens to the reflection of the music off neighboring surfaces, senses where the bookshelf is, or if it’s in the corner of a room or against a wall, and then uses machine learning to understand what it’s hearing, interpret the sound, and adjust the audio. Using advanced software and the A8 chip, HomePod intelligently beams center vocals and direct energy away from the wall while reflecting the ambient reverb and back-up vocals against the wall for dispersion into the room. The end result is a wide soundstage with a feeling of spaciousness and depth.
Talking about Siri on the HomePod, Schiller confirmed that Apple had made improvements to its virtual assistant so that it offers a more personalized experience on the smart speaker. Thanks to the tight integration with Apple Music, one can simply say “I like this song” to like a particular song being played on your HomePod. Similarly, one can also tell Siri to “play more like this” and Siri will automatically understand that the context is relevant to the currently played music. Apple also went to great lengths to ensure that “Hey Siri” on the HomePod works even when the speaker is blasting music at full volume. Schiller says that this is a great example of the collaboration between Apple’s software and hardware team.
HomePod uses advanced machine learning techniques including deep neural networks (DNNs) optimized for the hardware to detect “Hey Siri” in challenging environments. The detection of “Hey Siri” happens on the device, so nothing is being sent to Apple until that trigger is detected and the Siri waveform lights up. At that time, the request is sent to Apple using an anonymous Siri ID, and of course that communication is all encrypted. We’re able to do this because we designed audio technology, advanced software, and Siri to work together as one system.
Schiller also revealed during the interview that Siri now responds to over 2 billion requests every week.
Schiller talks about other aspects of the HomePod in the interview which might be worth checking out if you are looking to order a HomePod for yourself. The smart speaker is already up for pre-order in the United States, the United Kingdom, and Australia, with a release date of February 9.