Abstract: A method includes receiving training data including transcribed speech utterances spoken in a general domain, modified speech utterances in a target domain, and unspoken textual utterances corresponding to the transcriptions of the modified speech utterances in the target domain. The modified speech utterances include utterances spoken in the target domain that have been modified to obfuscate one or more classes of sensitive information recited in the utterances. The method also includes generating a corresponding alignment output for each unspoken textual utterance of the received training data using an alignment model. The method also includes training a speech recognition model on the alignment outputs generated for the corresponding to the unspoken textual utterances, the un-transcribed speech utterances, and the transcribed speech utterances to teach the speech recognition model to learn to recognize speech in the target domain and phrases within the one or more classes of sensitive information.

Abstract: An impact is detected with a vehicle having a body comprising parts which are deformable in an impact, by a secondary impact detection system configured to detect an impact that may not trigger a primary impact detection system. The secondary impact detection system comprises a plurality of deformation sensors attached to the deformable parts and configured to detect bending of the deformable parts. One or more processors are configured to receive a plurality of respective outputs from the plurality of deformation sensors, process the respective outputs to increase sensitivity to short-term changes relative to long-term changes and generate an indication of an impact dependent on detecting a processed output meeting a predetermined criterion. Outputs associated with a group of neighboring flex sensors may be used to localize the impact, characterize the source of impact, and/or reduce a likelihood of a false positive associated with the detection.

Abstract: The present technology can detect potential labeling conflicts made during different labeling tasks when those tasks are checked in to ensure that a map database is free from conflicts that might interrupt publishing of a map. The present technology can determine whether a first labeled version of map portion is based on a most recent version for the map portion that is stored in the map database, and when it is determined that it is not, and there is an intervening version, the present technology can identify differences between the intervening version of the map portion and the first version of the map portion and identify differences between the second version of the map portion and the first version of the map portion. Based on the identified differences, the present technology can determine whether any of those differences are conflicting.

Abstract: The present technology pertains to visualization of maps used for autonomous vehicle applications. These maps include many layers of dense information. Each layer can be stored in a vector format and small portions, called tiles, of a map can be requested by a users web browser. The web browser can filter for desired layers and render all desired map layers at once to provide visualization of map and layers of interest.

Abstract: The present technology pertains to visualization of maps used for autonomous vehicle applications. These maps include many layers of dense information. Each layer can be stored in a vector format and small portions, called tiles, of a map can be requested by a users web browser. The web browser can filter for desired layers and render all desired map layers at once to provide visualization of map and layers of interest.

Abstract: The present technology pertains to a task managment system for map labeling tasks. The system can determine when one task is dependent on completion of another task and schedule those tasks accordingly. Further the system can ensure that tasks are completed with a proper priority and can track completion of these tasks. The present technology ensures sufficent work is completed to speed up map release versions.

Abstract: The present technology pertains to a task management system for map labeling tasks. The system can determine when one task is dependent on completion of another task and schedule those tasks accordingly. Further the system can ensure that tasks are completed with a proper priority and can track completion of these tasks. The present technology ensures sufficient work is completed to speed up map release versions.

Abstract: The present technology pertains to a task management system for map labeling tasks. The system can determine when one task is dependent on completion of another task and schedule those tasks accordingly. Further the system can ensure that tasks are completed with a proper priority and can track completion of these tasks. The present technology ensures sufficient work is completed to speed up map release versions.

Abstract: The present technology can detect potential labeling conflicts made during different labeling tasks when those tasks are checked in to ensure that a map database is free from conflicts that might interrupt publishing of a map. The present technology can determine whether a first labeled version of map portion is based on a most recent version for the map portion that is stored in the map database, and when it is determined that it is not, and there is an intervening version, the present technology can identify differences between the intervening version of the map portion and the first version of the map portion and identify differences between the second version of the map portion and the first version of the map portion. Based on the identified differences, the present technology can determine whether any of those differences are conflicting.

Abstract: The present technology pertains to visualization of maps used for autonomous vehicle applications. These maps include many layers of dense information. Each layer can be stored in a vector format and small portions, called tiles, of a map can be requested by a users web browser. The web browser can filter for desired layers and render all desired map layers at once to provide visualization of map and layers of interest.

Abstract: The present technology pertains to visualization of maps used for autonomous vehicle applications. These maps include many layers of dense information. Each layer can be stored in a vector format and small portions, called tiles, of a map can be requested by a users web browser. The web browser can filter for desired layers and render all desired map layers at once to provide visualization of map and layers of interest.