Abstract: Systems and processes for operating a virtual assistant to provide natural assistant interaction are provided. In accordance with one or more examples, a method includes, at an electronic device with one or more processors and memory: receiving a first audio stream including one or more utterances; determining whether the first audio stream includes a lexical trigger; generating one or more candidate text representations of the one or more utterances; determining whether at least one candidate text representation of the one or more candidate text representations is to be disregarded by the virtual assistant. If at least one candidate text representation is to be disregarded, one or more candidate intents are generated based on candidate text representations of the one or more candidate text representations other than the to be disregarded at least one candidate text representation.

Abstract: Systems and processes for operating a virtual assistant to provide natural assistant interaction are provided. In accordance with one or more examples, a method includes, at an electronic device with one or more processors and memory: receiving a first audio stream including one or more utterances; determining whether the first audio stream includes a lexical trigger; generating one or more candidate text representations of the one or more utterances; determining whether at least one candidate text representation of the one or more candidate text representations is to be disregarded by the virtual assistant. If at least one candidate text representation is to be disregarded, one or more candidate intents are generated based on candidate text representations of the one or more candidate text representations other than the to be disregarded at least one candidate text representation.

Abstract: In some implementations, a computing device can manage conflicts using conflict islands. For example, when a computing device receives bulk map data updates, the computing device can determine conflicts between point updates (e.g., small, ad hoc feature edits) in the current map data and the bulk map data updates. When the bulk map data updates are merged with the current map data, a point update for a map feature may conflict with the bulk update for the feature. The computing device can determine a perimeter of stable map features around the conflicted feature. The map features within the perimeter can correspond to a conflict island. The bulk updates for features within a conflict island (e.g., within the perimeter) can be deferred until resolved while bulk updates for features outside of a conflict island can be applied to the current map data.

Abstract: In some implementations, a computing device can manage conflicts using conflict islands. For example, when a computing device receives bulk map data updates, the computing device can determine conflicts between point updates (e.g., small, ad hoc feature edits) in the current map data and the bulk map data updates. When the bulk map data updates are merged with the current map data, a point update for a map feature may conflict with the bulk update for the feature. The computing device can determine a perimeter of stable map features around the conflicted feature. The map features within the perimeter can correspond to a conflict island. The bulk updates for features within a conflict island (e.g., within the perimeter) can be deferred until resolved while bulk updates for features outside of a conflict island can be applied to the current map data.

Abstract: Systems and processes for operating a virtual assistant to provide natural assistant interaction are provided. In accordance with one or more examples, a method includes, at an electronic device with one or more processors and memory: receiving a first audio stream including one or more utterances; determining whether the first audio stream includes a lexical trigger; generating one or more candidate text representations of the one or more utterances; determining whether at least one candidate text representation of the one or more candidate text representations is to be disregarded by the virtual assistant. If at least one candidate text representation is to be disregarded, one or more candidate intents are generated based on candidate text representations of the one or more candidate text representations other than the to be disregarded at least one candidate text representation.

Abstract: In some implementations, a computing device can manage conflicts using conflict islands. For example, when a computing device receives bulk map data updates, the computing device can determine conflicts between point updates (e.g., small, ad hoc feature edits) in the current map data and the bulk map data updates. When the bulk map data updates are merged with the current map data, a point update for a map feature may conflict with the bulk update for the feature. The computing device can determine a perimeter of stable map features around the conflicted feature. The map features within the perimeter can correspond to a conflict island. The bulk updates for features within a conflict island (e.g., within the perimeter) can be deferred until resolved while bulk updates for features outside of a conflict island can be applied to the current map data.

Abstract: Systems and processes for operating a virtual assistant to provide natural assistant interaction are provided. In accordance with one or more examples, a method includes, at an electronic device with one or more processors and memory: receiving a first audio stream including one or more utterances; determining whether the first audio stream includes a lexical trigger; generating one or more candidate text representations of the one or more utterances; determining whether at least one candidate text representation of the one or more candidate text representations is to be disregarded by the virtual assistant. If at least one candidate text representation is to be disregarded, one or more candidate intents are generated based on candidate text representations of the one or more candidate text representations other than the to be disregarded at least one candidate text representation.

Abstract: Systems and processes for operating a virtual assistant to provide natural assistant interaction are provided. In accordance with one or more examples, a method includes, at an electronic device with one or more processors and memory: receiving a first audio stream including one or more utterances; determining whether the first audio stream includes a lexical trigger; generating one or more candidate text representations of the one or more utterances; determining whether at least one candidate text representation of the one or more candidate text representations is to be disregarded by the virtual assistant. If at least one candidate text representation is to be disregarded, one or more candidate intents are generated based on candidate text representations of the one or more candidate text representations other than the to be disregarded at least one candidate text representation.

Abstract: In some implementations, a computing device can manage conflicts using conflict islands. For example, when a computing device receives bulk map data updates, the computing device can determine conflicts between point updates (e.g., small, ad hoc feature edits) in the current map data and the bulk map data updates. When the bulk map data updates are merged with the current map data, a point update for a map feature may conflict with the bulk update for the feature. The computing device can determine a perimeter of stable map features around the conflicted feature. The map features within the perimeter can correspond to a conflict island. The bulk updates for features within a conflict island (e.g., within the perimeter) can be deferred until resolved while bulk updates for features outside of a conflict island can be applied to the current map data.

Abstract: Some embodiments provide a method for generating intersection data for paths in a map region. The method receives a set of junctions at which paths intersect in the map region. For a particular junction of at least two paths, the method automatically determines whether any of the other junctions in the map region satisfy criteria to be part of a single intersection with the particular junction. When at least one of the other junctions satisfies the criteria, the method automatically combines the other junctions that satisfy the criteria with the particular junction into a single intersection for use in performing mapping operations.

Abstract: Road height values for a road network area are generated based on locations of junctions and stacking orders of roads in junctions in which at least one road passes over another road. A cost function can be optimized to obtain height values for each road of each junction. The cost function can be constrained such that roads of the grade separated crossings satisfy the stacking order designated for the roads and to have a minimum distance between roads having consecutive stacking order values. The obtained height values can be used for rendering a three-dimensional map of a part of the road network area.

Abstract: Some embodiments provide a method for generating intersection data for paths in a map region. The method receives a set of junctions at which paths intersect in the map region. For a particular junction of at least two paths, the method automatically determines whether any of the other junctions in the map region satisfy criteria to be part of a single intersection with the particular junction. When at least one of the other junctions satisfies the criteria, the method automatically combines the other junctions that satisfy the criteria with the particular junction into a single intersection for use in performing mapping operations.

Abstract: Some embodiments provide a method for generating intersection data for paths in a map region. The method receives a set of junctions at which paths intersect in the map region. For a particular junction of at least two paths, the method automatically determines whether any of the other junctions in the map region satisfy criteria to be part of a single intersection with the particular junction. When at least one of the other junctions satisfies the criteria, the method automatically combines the other junctions that satisfy the criteria with the particular junction into a single intersection for use in performing mapping operations.

Abstract: Road height values for a road network area are generated based on locations of junctions and stacking orders of roads in junctions in which at least one road passes over another road. A cost function can be optimized to obtain height values for each road of each junction. The cost function can be constrained such that roads of the grade separated crossings satisfy the stacking order designated for the roads and to have a minimum distance between roads having consecutive stacking order values. The obtained height values can be used for rendering a three-dimensional map of a part of the road network area.

Abstract: Some embodiments provide a method for generating intersection data for paths in a map region. The method receives a set of junctions at which paths intersect in the map region. For a particular junction of at least two paths, the method automatically determines whether any of the other junctions in the map region satisfy criteria to be part of a single intersection with the particular junction. When at least one of the other junctions satisfies the criteria, the method automatically combines the other junctions that satisfy the criteria with the particular junction into a single intersection for use in performing mapping operations.