Component libraries for voice interaction services

One or more non-transitory computer-readable media having instructions stored thereon that, when executed by data processing apparatus of a computing device, cause the data processing apparatus to perform operations comprising:

• receiving, by the computing device and with a first client application installed on the computing device, an utterance that was spoken by a user of the computing device;
• processing the utterance with the computing device to generate a voice query, the voice query including data that is based on the utterance;
• transmitting the voice query from the computing device to a remote computing system associated with a voice service provider application installed on the computing device, wherein the voice service provider application is separate from the first client application;
• obtaining, by the computing device and from the remote computing system associated with the voice service provider application, a response to the voice query;
• selecting, by the first client application, a user interface element from a library of user interface elements associated with the voice service provider application;
• and
• presenting the response to the voice query at the computing device,
• wherein the user interface element from the library associated with the voice service provider is embedded into a native user interface of the first client application,
• wherein the library associated with the voice service provider application comprises a library of application-neutral elements of executable code that, when executed, embed user interface elements into native user interfaces of a plurality of applications executed by the computing device, and
• wherein the embedded user interface elements are selectable to initiate, by the voice service provider application, voice-based interactions with the native user interfaces of the plurality of applications executed by the computing device.