Abstract: A method of generating diverse and natural text-to-speech (TTS) samples includes receiving a text and generating a speech sample based on the text using a TTS model. A training process trains the TTS model to generate the speech sample by receiving training samples. Each training sample includes a spectrogram and a training text corresponding to the spectrogram. For each training sample, the training process identifies speech units associated with the training text. For each speech unit, the training process generates a speech embedding, aligns the speech embedding with a portion of the spectrogram, extracts a latent feature from the aligned portion of the spectrogram, and assigns a quantized embedding to the latent feature. The training process generates the speech sample by decoding a concatenation of the speech embeddings and a quantized embeddings for the speech units associated with the training text corresponding to the spectrogram.

Abstract: Disclosed herein are polynucleotides encoding multi-epitope polypeptides and assemblies thereof, and their use for treating or limiting Toxoplasma gondii infection.

Abstract: Provided herein are genetic circuits and cell state classifiers for detecting the microRNA profile of a cell. The cell state classifiers of the present disclosure are designed to incorporate multiple genetic circuits integrated together by transcriptional or translational control. Multiple inputs can be sensed simultaneously by coupling their detection to different portions of the genetic circuit such that the output molecule is produced only when the correct input profile of miRNAs is detected. The genetic circuits and cell state classifiers may be used in various applications (e.g., therapeutic or diagnostic applications).

Abstract: Engineered synthetic RNA-based genetic circuits are provided that are regulated exclusively at the post-transcriptional level.

Abstract: Provided herein are genetic circuits and cell state classifiers for detecting the microRNA profile of a cell. In some embodiments, the cell state classifiers described herein utilize an endoribonuclease and a self-amplifying RNA molecule for controlling the expression of an output molecule. In some embodiments, the cell state classifiers described herein are encoded on a single RNA transcript, which is then processed to produce individual genetic circuits that function independently. The genetic circuits and cell state classifiers described herein may be used in various applications (e.g., therapeutic or diagnostic applications).

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for neural network adaptive beamforming for multichannel speech recognition are disclosed. In one aspect, a method includes the actions of receiving a first channel of audio data corresponding to an utterance and a second channel of audio data corresponding to the utterance. The actions further include generating a first set of filter parameters for a first filter based on the first channel of audio data and the second channel of audio data and a second set of filter parameters for a second filter based on the first channel of audio data and the second channel of audio data. The actions further include generating a single combined channel of audio data. The actions further include inputting the audio data to a neural network. The actions further include providing a transcription for the utterance.

Abstract: Compositions and methods are described for inducing an immune response against an immunogen in humans. The induced immune response is obtained by administering a heterologous prime-boost combination of an in vitro transcribed (IVT) self-replicating RNA (repRNA) and an adenovirus vector. The compositions and methods can be used to provide a protective immunity against a disease, such as a viral infection or a cancer, in humans.

Abstract: Compositions and methods are described for inducing an immune response against an immunogen in humans. The induced immune response is obtained by administering a heterologous prime-boost combination of an in vitro transcribed (IVT) self-replicating RNA (repRNA) and an adenovirus vector. The compositions and methods can be used to provide a protective immunity against a disease, such as a viral infection or a cancer, in humans.

Abstract: Disclosed herein are genetically modified herpesviruses for the treatment of cancer. Also provided are methods of treating cancer using genetically modified herpesviruses.

Abstract: The present disclosure, at least in part, relates to an engineered RNA (e.g., microRNA and sgRNA), in the absence of an input signal, that is engineered to have a large enough energy gap between the formations of a first secondary structure, which is unrecognizable by an actuator, and a second secondary structure, which is recognizable by an actuator (e.g., Drosha and Cas protein).

Abstract: Genetic circuits have been developed to regulate behaviors of replicon RNA in responses to small molecules, which has broader applications, such as for quantitative expression of cargo genes, temporary expression of immunomodulatory cytokines or antigens for better cancer immunotherapy or vaccination, and for increased safety in use of self-replicating vectors or in combination with other viral-delivery vectors. Described herein are genetic circuits suitable for systems that either require a tight off state or a slow off state, which can serve for instance where either a kill switch or prolonged protein expression (e.g., of vaccine antigens) are needed.

Abstract: This disclosure provides engineered genetic systems for sequestering and/or destroying viruses, compositions and cells comprising the genetic systems, and methods of treating viral infections, reducing viral load, and/or reducing viral spread. The disclosure also provides libraries comprising elements to be incorporated into the engineered genetic systems.

Abstract: The present disclosure, at least in part, relates to an engineered incoherent feed forward loop (iFFL) comprising a first transcription unit encoding an endoribonuclease and a second transcription unit encoding an output molecule and an endoribonuclease target site located in the 5? UTR of the output molecule coding sequence. The engineered iFFL, at least in part, can be used for sustained expression of an output molecule despite of transcription resource disturbance in a given environment.

Abstract: The present disclosure, at least in part, provides RNA cleavage based engineered bi-stable toggle switch utilizing the Programmable Endonucleolytic Scission-Induced Stability Tuning (PERSIST) platform. Also provided herein, are vectors encoding the engineered bi-stable toggle switch, and uses thereof.

Abstract: A method includes receiving a training example for a listen-attend-spell (LAS) decoder of a two-pass streaming neural network model and determining whether the training example corresponds to a supervised audio-text pair or an unpaired text sequence. When the training example corresponds to an unpaired text sequence, the method also includes determining a cross entropy loss based on a log probability associated with a context vector of the training example. The method also includes updating the LAS decoder and the context vector based on the determined cross entropy loss.

Abstract: The present disclosure provides systems and methods that train and use machine-learned models such as, for example, sequence-to-sequence models, to perform direct and text-free speech-to-speech translation. In particular, aspects of the present disclosure provide an attention-based sequence-to-sequence neural network which can directly translate speech from one language into speech in another language, without relying on an intermediate text representation.

Abstract: Disclosed herein are lipid nanoparticles comprising 1,2-dioleoyl-3-trimethylammonium propane (DOTAP), 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), cholesterol, and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino(polyethylene glycol)-2000](DSPE-PEG2000), compositions including such lipid nanoparticles, and methods of treatment using such lipid nanoparticles and compositions.

Abstract: Disclosed herein are methods of generating proteoglycans with distinct glycan structures in engineered, non-naturally occurring eukaryotic cells. These methods make accessible a dynamic range of protein glycosylation. Compositions of engineered, non-naturally occurring cells capable of generating these proteoglycans are also disclosed herein.

Abstract: The present disclosure, at least in part, relates to a miRNA based logic gate that comprises an engineered RNA carrier that comprises an nuclear export signal, a target site for a first miRNA and a pre-miRNA sequence for a second miRNA. Also provided by the disclosure are recombinant viruses (e.g., recombinant adeno-associated viruses (rAAV)) for delivery of the miRNA based logic gates.

Abstract: The present disclosure, at least in part, relates to an engineered RNA (e.g., microRNA and sgRNA), in the absence of an input signal, that is engineered to have a large enough energy gap between the formations of a first secondary structure, which is unrecognizable by an actuator, and a second secondary structure, which is recognizable by an actuator (e.g., Drosha and Cas protein).