Abstract: In various examples, systems and methods that use dialogue systems associated with various machine systems and applications are described. For instance, the systems and methods may receive text data representing speech, such as a question associated with a vehicle or other machine type. The systems and methods then use a retrieval system(s) to retrieve a question/answer pair(s) associated with the text data and/or contextual information associated with the text data. In some examples, the contextual information is associated with a knowledge base associated with or corresponding to the vehicle. The systems and methods then generate a prompt using the text data, the question/answer pair(s), and/or the contextual information. Additionally, the systems and methods determine, using a language model(s) and based at least on the prompt, an output associated with the text data. For instance, the output may include information that answers the question associated with the vehicle.

Abstract: Apparatuses, systems, and techniques are presented to identify and prevent generation of restricted content. In at least one embodiment, one or more neural networks are used to identify restricted content based only on the restricted content.

Abstract: Embodiments of a production-quality text-to-speech (TTS) system constructed from deep neural networks are described. System embodiments comprise five major building blocks: a segmentation model for locating phoneme boundaries, a grapheme-to-phoneme conversion model, a phoneme duration prediction model, a fundamental frequency prediction model, and an audio synthesis model. For embodiments of the segmentation model, phoneme boundary detection was performed with deep neural networks using Connectionist Temporal Classification (CTC) loss. For embodiments of the audio synthesis model, a variant of WaveNet was created that requires fewer parameters and trains faster than the original. By using a neural network for each component, system embodiments are simpler and more flexible than traditional TTS systems, where each component requires laborious feature engineering and extensive domain expertise. Inference with system embodiments may be performed faster than real time.

Abstract: Described herein are systems and methods for compressing and speeding up dense matrix multiplications as found, for examples, in the fully connected and recurrent layers of neural networks for embedded large vocabulary continuous speech recognition (LVCSR). For compression, trace norm regularization technique embodiments were introduced and studied for training low rank factored versions of matrix multiplications. Compared to standard low rank training, the methods more consistently lead to good accuracy versus number of parameter trade-offs and can be used to speed-up training of large models. Faster inference may be further enabled on ARM processors through kernels optimized for small batch sizes, resulting in speed ups over the currently used library. Beyond LVCSR, the techniques are also generally applicable to embedded neural networks with large fully connected or recurrent layers.

Abstract: Apparatuses, systems, and techniques to enhance video. In at least one embodiment, one or more neural networks are used to create, from a first video, a second video having a higher frame rate, higher resolution, or reduced number of missing or corrupt video frames.

Abstract: Embodiments of a production-quality text-to-speech (TTS) system constructed from deep neural networks are described. System embodiments comprise five major building blocks: a segmentation model for locating phoneme boundaries, a grapheme-to-phoneme conversion model, a phoneme duration prediction model, a fundamental frequency prediction model, and an audio synthesis model. For embodiments of the segmentation model, phoneme boundary detection was performed with deep neural networks using Connectionist Temporal Classification (CTC) loss. For embodiments of the audio synthesis model, a variant of WaveNet was created that requires fewer parameters and trains faster than the original. By using a neural network for each component, system embodiments are simpler and more flexible than traditional TTS systems, where each component requires laborious feature engineering and extensive domain expertise. Inference with system embodiments may be performed faster than real time.

Abstract: Embodiments of a production-quality text-to-speech (TTS) system constructed from deep neural networks are described. System embodiments comprise five major building blocks: a segmentation model for locating phoneme boundaries, a grapheme-to-phoneme conversion model, a phoneme duration prediction model, a fundamental frequency prediction model, and an audio synthesis model. For embodiments of the segmentation model, phoneme boundary detection was performed with deep neural networks using Connectionist Temporal Classification (CTC) loss. For embodiments of the audio synthesis model, a variant of WaveNet was created that requires fewer parameters and trains faster than the original. By using a neural network for each component, system embodiments are simpler and more flexible than traditional TTS systems, where each component requires laborious feature engineering and extensive domain expertise. Inference with system embodiments may be performed faster than real time.

Abstract: Described herein are systems and methods for compressing and speeding up dense matrix multiplications as found, for examples, in the fully connected and recurrent layers of neural networks for embedded large vocabulary continuous speech recognition (LVCSR). For compression, trace norm regularization technique embodiments were introduced and studied for training low rank factored versions of matrix multiplications. Compared to standard low rank training, the methods more consistently lead to good accuracy versus number of parameter trade-offs and can be used to speed-up training of large models. Faster inference may be further enabled on ARM processors through kernels optimized for small batch sizes, resulting in speed ups over the currently used library. Beyond LVCSR, the techniques are also generally applicable to embedded neural networks with large fully connected or recurrent layers.

Abstract: Embodiments of a production-quality text-to-speech (TTS) system constructed from deep neural networks are described. System embodiments comprise five major building blocks: a segmentation model for locating phoneme boundaries, a grapheme-to-phoneme conversion model, a phoneme duration prediction model, a fundamental frequency prediction model, and an audio synthesis model. For embodiments of the segmentation model, phoneme boundary detection was performed with deep neural networks using Connectionist Temporal Classification (CTC) loss. For embodiments of the audio synthesis model, a variant of WaveNet was created that requires fewer parameters and trains faster than the original. By using a neural network for each component, system embodiments are simpler and more flexible than traditional TTS systems, where each component requires laborious feature engineering and extensive domain expertise. Inference with system embodiments may be performed faster than real time.