Abstract: A method includes, while a user is wearing stereo headphones in an environment, obtaining, from a target digital assistant, a response to a query issued by the user, and obtaining spatial audio preferences of the user. Based on the spatial audio preferences of the user, the method also includes determining a spatially disposed location within a playback sound-field for the user to perceive as a sound-source of the response to the query. The method further includes rendering output audio signals characterizing the response to the query through the stereo headphones to produce the playback sound-field. Here, the user perceives the response to the query as emanating from the sound-source at the spatially disposed location within the playback sound-field.

Abstract: A method includes receiving user queries directed toward a cloud-based assistant service. For each received user query directed toward the cloud-based assistant service, the method also includes extracting one or more attributes from the user query and logging the user query into one or more of a plurality of category buckets based on the one or more attributes extracted from the user query. The method also includes determining when at least one of the plurality of category buckets includes a threshold number of the user queries logged into the at least one category bucket, and when the at least one of the plurality of category buckets includes the threshold number of the user queries, generating a distilled model of the cloud-based assistant service. The distilled model of the cloud-based assistant service is configured to execute on one or more target client devices.

Abstract: Implementations set forth herein relate to an automated assistant that can operate in a transient personalization mode, and/or assist a separate automated assistant with providing output according to a transient personalization mode. The transient personalization mode can allow a guest user of an assistant enabled-device to receive personalized responses from the assistant-enabled device—despite not being signed into the assistant-enabled device. A host automated assistant of the assistant-enabled device can securely communicate with a guest user's automated assistant through a backend process. In this way, input queries from the guest user to the host automated assistant can be personalized according to the guest automated assistant—without the guest user directly engaging with their own personal device.

Abstract: A method for combining hotwords in a single utterance receives, at a first assistant-enabled device (AED), audio data corresponding to an utterance directed toward the first AED and a second AED among two or more AEDs where the audio data includes a query specifying an operation to perform. The method also detects, using a hotword detector, a first hotword assigned to the first AED that is different than a second hotword assigned to the second AED. In response to detecting the first hotword, the method initiates processing on the audio data to determine that the audio data includes a term preceding the query that at least partially matches the second hotword assigned. Based on the at least partial match, the method executes a collaboration routine to cause the first AED and the second AED to collaborate with one another to fulfill the query.

Abstract: Systems and methods for determining, based on invocation input that is common to multiple automated assistants, which automated assistant to invoke in lieu of invoking other automated assistants. The invocation input is processed to determine one or more invocation features that may be utilized to determine which, of a plurality of candidate automated assistants, to invoke. Further, additional features are processed that can indicate which, of the plurality of invocable automated assistants, to invoke. Once an automated assistant has been invoked, additional audio data and/or features of additional audio data are provided to the invoked automated assistant for further processing.

Abstract: A computing device is disclosed that includes at least one processor and a machine-learned model. The machine-learned model can include a plurality of blocks and one or more residual connections between two or more of the plurality of blocks. The machine-learned model can be configured to receive a model input and, in response to receipt of the model input, output a model output. The machine-learned model can be configured to perform operations including determining a resource allocation parameter that corresponds to a desired allocation of system resources to the machine-learned model at an inference time: deactivating a subset of the plurality of blocks of the machine-learned model based on the resource allocation parameter: inputting the model input into the machine-learned model with the subset of the plurality of blocks deactivated; and receiving, as an output of the machine-learned model, the model output.

Abstract: While an assistant-enabled device is playing back media content, a method includes receiving a contextual signal from an environment of the assistant-enabled device and executing an event recognition routine to determine whether the received contextual signal is indicative of an event that conflicts with the playback of the media content from the assistant-enabled device. When the event recognition routine determines that the received contextual signal is indicative of the event that conflicts with the playback of the media content, the method also includes adjusting content playback settings of the assistant-enabled device.

Abstract: Systems and methods for providing dialog data, from an initially invoked automated assistant to a subsequently invoked automated assistant. A first automated assistant may be invoked by a user utterance, followed by a dialog with the user that is processed by the first automated assistant. During the dialog, a request to transfer dialog data to a second automated assistant is received. The request may originate with the user, by the first automated assistant, and/or by the second automated assistant. Once authorized, the first automated assistant provides the previous dialog data to the second automated assistant. The second automated assistant performs one or more actions based on the dialog data.

Abstract: Implementations relate to interactions between a user and an automated assistant during a dialog between the user and the automated assistant. Some implementations relate to processing received user request input to determine that it is of a particular type that is associated with a source parameter rule and, in response, causing one or more sources indicated as preferred by the source parameter rule and one or more additional sources not indicated by the source parameter rule to be searched based on the user request input. Further, those implementations relate to identifying search results of the search(es), and generating, in dependence on the search results, a response to the user request that includes content from search result(s) of the preferred source(s) and/or content from search result(s) of the additional source(s). Generating the response further includes including, in the response, some indication that indicates whether the source parameter rule was followed or violated in generating the response.

Abstract: Implementations set forth herein relate to an automated assistant that provides a response for certain user queries based on a level of interaction of the user with respect to the automated assistant. Interaction can be characterized by sensor data, which can be processed using one or more trained machine learning models in order to identify parameters for generating a response. In this way, the response can be limited to preserve computational resources and/or ensure that the response is more readily understood given the amount of interaction exhibited by the user. In some instances, a response that embodies information that is supplemental, to an otherwise suitable response, can be provided when a user is exhibiting a particular level of interaction. In other instances, such supplemental information can be withheld when the user is not exhibiting that particular level of interaction, at least in order to preserve computational resources.

Abstract: Implementations related to facilitating continued conversations of a user with an automated assistant when the user changes locations relative to one or more devices in an ecosystem of linked assistant devices. The user initially invokes a first device and provides a request, which is processed by the first device. The first device provides a notification to one or more other devices in the ecosystem to indicate that the user is likely to issue a further assistant request. The first device processes subsequent audio data to determine whether the subsequent audio data includes a further assistant request. The one or more other notified devices process device-specific sensor data to determine whether the user is co-present with the one of the other devices. If the user presence is detected, an indication is provided to the first device, causing the first device to cease processing subsequent audio data. Further, the co-present device starts to process subsequent audio data.

Abstract: Systems and methods for generating and providing outputs in a multi-device system can include leveraging environment-based prompt generation and generative model response generation to provide dynamic response generation and display. The systems and methods can obtain input data associated with one or more computing devices within an environment, can obtain environment data descriptive of the plurality of computing devices within the environment, and can generate a prompt based on the input data and environment data. The prompt can be processed with a generative model to generate a model-generated output. The model-generated output can then be transmitted to a particular computing device of the plurality of computing devices.

Abstract: A method for providing proactive assistance includes obtaining, by a digital assistant, a contextual event associated with a user of a user device. The method includes determining, using a local large language model (LLM) executing on the user device, a remote LLM prompt confidence. The method includes determining that the remote LLM prompt confidence satisfies a threshold. Based on determining that the remote LLM prompt confidence satisfies the threshold, the method includes generating a remote LLM prompt for a remote LLM executing remote from the user device. The method includes transmitting, to the remote LLM, the remote LLM prompt. The method includes receiving, at the digital assistant, from the remote LLM, response content providing the proactive assistance associated with the contextual event. The method includes providing, for output from the user device, presentation content based on the response content received from the remote LLM.

Abstract: A method for personalizing responses includes receiving a natural language query from a user that solicits a response from an assistant large language model (LLM) and processing the natural language query to identify, from a log of refinement queries each previously input by the user for instructing the assistant LLM to refine a corresponding previous response generated by the assistant LLM, a preferred refinement query that is relevant to the natural language query. The method also includes prompting, using the natural language query and the preferred refinement query identified as being relevant to the natural language query, the assistant LLM to generate a personalized response to the natural language query. The method also includes providing, for output from a user device, the personalized response to the natural language query.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for obtaining, for each of multiple words or sub-words, audio data corresponding to multiple users speaking the word or sub-word; training, for each of the multiple words or sub-words, a pre-computed hotword model for the word or sub-word based on the audio data for the word or sub-word; receiving a candidate hotword from a computing device; identifying one or more pre-computed hotword models that correspond to the candidate hotword; and providing the identified, pre-computed hotword models to the computing device.

Abstract: A method for decaying speech processing includes receiving, at a voice-enabled device, an indication of a microphone trigger event indicating a possible interaction with the device through speech where the device has a microphone that, when open, is configured to capture speech for speech recognition. In response to receiving the indication of the microphone trigger event, the method also includes instructing the microphone to open or remain open for a duration window to capture an audio stream in an environment of the device and providing the audio stream captured by the open microphone to a speech recognition system. During the duration window, the method further includes decaying a level of the speech recognition processing based on a function of the duration window and instructing the speech recognition system to use the decayed level of speech recognition processing over the audio stream captured by the open microphone.

Abstract: Information indicative of a particular geographic area in which a user computing device is located is obtained, wherein the particular geographic area comprises a plurality of Points of Interest (POIs). For a first POI of the plurality of POIs, a first set of textual content is generated with a first machine-learned POI-specific language model configured to generate textual content from a perspective of the first POI. For a second POI of the plurality of POIs, a second set of textual content is generated with a second machine-learned POI-specific language model configured to generate textual content from a perspective of the second POI. At least some of the second set of textual content is provided to the user computing device.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for hotword detection on multiple devices are disclosed. In one aspect, a method includes the actions of receiving, by a first computing device, audio data that corresponds to an utterance. The actions further include determining a first value corresponding to a likelihood that the utterance includes a hotword. The actions further include receiving a second value corresponding to a likelihood that the utterance includes the hotword, the second value being determined by a second computing device. The actions further include comparing the first value and the second value. The actions further include based on comparing the first value to the second value, initiating speech recognition processing on the audio data.

Abstract: Methods, systems, and computer readable media related to generating a combined search query based on search parameters of a current search query of a user and search parameters of one or more previously submitted search quer(ies) of the user that are determined to be of the same line of inquiry as the current search query. Two or more search queries may be determined to share a line of inquiry when it is determined that they are within a threshold level of semantic similarity to one another. Once a shared line of inquiry has been identified and a combined search query generated, users may interact with the search parameters and/or the search results to update the search parameters of the combined search query.

Abstract: Techniques are described herein for providing smart suggestions for image zoom regions. A method includes: receiving a search query; performing a search using the search query to identify search results that include image search results including a plurality of images that are responsive to the search query; for a given image of the plurality of images included in the image search results, determining at least one zoom region in the given image; and providing the search results including the image search results, including providing the given image and an indication of the at least one zoom region in the given image.