摘要:
According to one embodiment, a method, apparatus for synthesizing speech, and a method for training acoustic model used in speech synthesis is provided. The method for synthesizing speech may include determining data generated by text analysis as fuzzy heteronym data, performing fuzzy heteronym prediction on the fuzzy heteronym data to output a plurality of candidate pronunciations of the fuzzy heteronym data and probabilities thereof, generating fuzzy context feature labels based on the plurality of candidate pronunciations and probabilities thereof, determining model parameters for the fuzzy context feature labels based on acoustic model with fuzzy decision tree, generating speech parameters from the model parameters, and synthesizing the speech parameters via synthesizer as speech.
摘要:
According to one embodiment, a method, apparatus for synthesizing speech, and a method for training acoustic model used in speech synthesis is provided. The method for synthesizing speech may include determining data generated by text analysis as fuzzy heteronym data, performing fuzzy heteronym prediction on the fuzzy heteronym data to output a plurality of candidate pronunciations of the fuzzy heteronym data and probabilities thereof, generating fuzzy context feature labels based on the plurality of candidate pronunciations and probabilities thereof, determining model parameters for the fuzzy context feature labels based on acoustic model with fuzzy decision tree, generating speech parameters from the model parameters, and synthesizing the speech parameters via synthesizer as speech.
摘要:
According to one embodiment, an apparatus for synthesizing a speech, comprises an inputting unit configured to input a text sentence, a text analysis unit configured to analyze the text sentence so as to extract linguistic information, a parameter generation unit configured to generate a speech parameter by using the linguistic information and a pre-trained statistical parameter model, an embedding unit configured to embed information into the speech parameter, and a speech synthesis unit configured to synthesize the speech parameter with the information embedded by the embedding unit into a speech with the information.
摘要:
An ice cube tray box for facilitating ice making and ice removing, which includes a box body, a cover body arranged at an open end of the box body, and an ice cube tray body arranged inside the box body, wherein a vertical upward rib assembly is arranged on an inner wall of the box body, a notch assembly matched with the rib assembly in use is arranged on an outer edge of the ice cube tray body, an ice removing plate for facilitating ice removing is clamped and arranged on the inner side of the cover body, and ice knocking columns are integrally arranged on the ice removing plate.
摘要:
The present invention is a novel baby holder with a main body configured having a central cavity constructed and arrange to hold a baby with the main cavity having a front pair of retractable legs and a rear pair of retractable legs.
摘要:
An organic light emitting device including an emissive layer including a blue phosphorescent emitter, an electron transport layer, and a hole blocking layer between the emissive layer and the electron transport layer, wherein the hole blocking layer comprises a tetradentate palladium complex.
摘要:
A valve for m exhaust gas recirculation Mae in an engine includes a first barrel connectable to a first group of cylinders, a second barrel connectable to a second group of cylinders, and a center barrel between the first and second barrels and connectable to an intake of the engine, A poppet valve with two valve heads mounted on a valve stem is provided to open and close openings between the first and second barrels, with one of the valve beads of the poppet valve opening into the center barrel Vanes are provided in the center barrel to direct flow from the opening by the valve bead that opens into the center barrel toward the outlet opening.
摘要:
Various methods and communications devices to reduce a bandwidth utilization of a backhaul link in a wireless communications system are provided. By way of example, bandwidth utilization is reduced by relegating the compression of data to the downlink transmission, storing only reference keys in the cache of the transmitting device, and taking advantage of an asymmetrical cache structure between communicating devices. Support is provided for a user equipment to move from one node to another node.
摘要:
A trench metal-oxide-semiconductor field effect transistor (MOSFET), in accordance with one embodiment, includes a drain region, a plurality of gate regions disposed above the drain region, a plurality of gate insulator regions each disposed about a periphery of a respective one of the plurality of gate regions, a plurality of source regions disposed in recessed mesas between the plurality of gate insulator regions, a plurality of body regions disposed in recessed mesas between the plurality of gate insulator regions and between the plurality of source regions and the drain region. The MOSFET also includes a plurality of body contact regions disposed in the each body region adjacent the plurality of source regions, a plurality of source/body contact spacers disposed between the plurality of gate insulator regions above the recessed mesas, a source/body contact disposed above the source/body contact spacers, and a plurality of source/body contact plugs disposed between the source/body contact spacers and coupling the source/body contact to the plurality of body contact regions and the plurality of source regions.