254 results about "Audio synthesis" patented technology

Systems and methods for creating, modifying, interacting with and playing music are provided, particularly systems and methods employing a top-down process, where the user is provided with a musical composition that may be modified and interacted with and played and/or stored (for later play). The system preferably is provided in a handheld form factor, and a graphical display is provided to display status information, graphical representations of musical lanes or components which preferably vary in shape as musical parameters and the like are changed for particular instruments or musical components such as a microphone input or audio samples. An interactive auto-composition process preferably is utilized that employs musical rules and preferably a pseudo random number generator, which may also incorporate randomness introduced by timing of user input or the like, the user may then quickly begin creating desirable music in accordance with one or a variety of musical styles, with the user modifying the auto-composed (or previously created) musical composition, either for a real time performance and/or for storing and subsequent playback. In addition, an analysis process flow is described for using pre-existing music as input(s) to an algorithm to derive music rules that may be used as part of a music style in a subsequent auto-composition process. In addition, the present invention makes use of node-based music generation as part of a system and method to broadcast and receive music data files, which are then used to generate and play music. By incorporating the music generation process into a node-subscriber unit, the bandwidth-intensive systems of conventional techniques can be avoided. Consequently, the bandwidth can preferably be also used of additional features such as node-to-node and node to base music data transmission. The present invention is characterized by the broadcast of relatively small data files that contain various parameters sufficient to describe the music to the node/subscriber music generator. In addition, problems associated with audio synthesis in a portable environment are addressed in the present invention by providing systems and methods for performing audio synthesis in a manner that simplifies design requirements and/or minimizes cost, while still providing quality audio synthesis features targeted for a portable system (e.g., portable telephone). In addition, problems associated with the tradeoff between overall sound quality and memory requirements in a MIDI sound bank are addressed in the present invention by providing systems and methods for a reduced memory size footprint MIDI sound bank.

An audio synthesizing apparatus receives an encoded signal comprising a downmix signal and parametric extension data for expanding the downmix signal to a multi-sound source signal. A decomposition processor (205) performs a signal decomposition of the downmix signal to generate at least a first signal component and a second signal component, where the second signal component is at least partially decorrelated with the first signal component. A position processor (207) determines a first spatial position indication for the first signal component in response to the parametric extension data and a binaural processor (211) synthesizes the first signal component based on the first spatial position indication and the second signal component to originate from a different direction. The invention may provide improved spatial experience from e.g. headphones by using a direct synthesis of a main directional signal from the appropriate position rather than as a combination of signals from virtual loudspeaker positions.

The invention discloses an integrated sensor based on a microphone array and stereoscopic vision. The integrated sensor comprises a hardware system and a control system. The hardware system comprises the microphone array and a stereoscopic vision device. The microphone array is responsible for receiving audio information in an environment and generating multichannel analog audios. The stereoscopic vision device is responsible for collecting environment images. The control system comprises an audio working area, a video working area and a video/audio synthesis working area. The audio working area is responsible for denoising audios, positioning audio sources, and tracking the audio sources (through audio processing). The video working area is responsible for positioning and tracking target audio sources, namely calibrating, positioning and precisely tracking the target audio sources through image processing and image identification. The video/audio synthesis working area is responsible for enhancing audio signals (improving a signal-to-noise ratio of the audio signals) and obtaining and outputting an integrated signal of the video/audio signals.

Systems and methods for creating, modifying, interacting with and playing music are provided, preferably employing a top-down process, where the user is provided with a musical composition that may be modified and interacted with and played and/or stored (for later play). The system preferably is provided in a handheld form factor, and a graphical display is provided to display status information, graphical representations of musical lanes or components which preferably vary in shape as musical parameters and the like are changed for particular instruments or musical components such as a microphone input or audio samples. An interactive auto-composition process preferably employs musical rules and a pseudo random number generator, which may also incorporate randomness introduced by timing of user input or the like. The user may quickly begin creating desirable music in accordance with one or a variety of musical styles, with the user modifying the auto-composed (or previously created) musical composition, either for a real time performance and/or for storing and subsequent playback. An analysis process flow also is disclosed for using pre-existing music as input(s) to an algorithm to derive music rules that may be used as part of a music style in a subsequent auto-composition process. In addition, the present invention makes use of node-based music generation as part of a system and method to broadcast and receive music data files, which are then used to generate and play music. By incorporating the music generation process into a node/subscriber unit, bandwidth requirements are lowered, and consequently the bandwidth can preferably be used for additional features such as node-to-node and node-to-base music data transmission. The present invention is characterized by the broadcast of relatively small data files that contain various parameters sufficient to describe the music to the node/subscriber music generator. In addition, improved audio synthesis in a portable environment is provided with the present invention by performing audio synthesis in a manner that simplifies design requirements and/or minimizes cost, while still providing quality audio synthesis features targeted for a portable system (e.g., portable telephone). In addition, problems associated with the tradeoff between overall sound quality and memory requirements in a MIDI sound bank are addressed in the present invention by providing systems and methods for a reduced memory size footprint MIDI sound bank. In addition, music ringtone alert tone remixing, navigation, and purchasing capabilities are disclosed that are particularly advantageous in the context of a portable communications device, such as a cellular telephone, in connection with a communications network.

Systems and methods for creating, modifying, interacting with and playing music are provided, preferably employing a top-down process, where the user is provided with a musical composition that may be modified and interacted with and played and/or stored (for later play). The system preferably is provided in a handheld form factor, and a graphical display is provided to display status information, graphical representations of musical lanes or components which preferably vary in shape as musical parameters and the like are changed for particular instruments or musical components such as a microphone input or audio samples. An interactive auto-composition process preferably employs musical rules and a pseudo random number generator, which may also incorporate randomness introduced by timing of user input or the like. The user may quickly begin creating desirable music in accordance with one or a variety of musical styles, with the user modifying the auto-composed (or previously created) musical composition, either for a real time performance and/or for storing and subsequent playback. An analysis process flow also is disclosed for using pre-existing music as input(s) to an algorithm to derive music rules that may be used as part of a music style in a subsequent auto-composition process. In addition, the present invention makes use of node-based music generation as part of a system and method to broadcast and receive music data files, which are then used to generate and play music. By incorporating the music generation process into a node/subscriber unit, bandwidth requirements are lowered, and consequently the bandwidth can preferably be used for additional features such as node-to-node and node-to-base music data transmission. The present invention is characterized by the broadcast of relatively small data files that contain various parameters sufficient to describe the music to the node/subscriber music generator. In addition, improved audio synthesis in a portable environment is provided with the present invention by performing audio synthesis in a manner that simplifies design requirements and/or minimizes cost, while still providing quality audio synthesis features targeted for a portable system (e.g., portable telephone). In addition, problems associated with the tradeoff between overall sound quality and memory requirements in a MIDI sound bank are addressed in the present invention by providing systems and methods for a reduced memory size footprint MIDI sound bank. In addition, music ringtone alert tone remixing, navigation, and purchasing capabilities are disclosed that are particularly advantageous in the context of a portable communications device, such as a cellular telephone, in connection with a communications network.

A video stream decoder include: a demultiplexing unit for demultiplexing a video stream containing at least one or more object encoded visual or audio data and one or more scene descriptions which express scene contents by object encoded data; a decoder unit for decoding the object encoded visual data; a decoder unit for decoding the object encoded audio data; a visual synthesizing unit for synthesizing images corresponding to the object encoded visual data; an audio synthesizing unit for synthesizing sounds corresponding to the object encoded audio data; an analyzing unit for analyzing each scene description; a selector for selecting one of at least two or more scene descriptions contained in the video stream.

A selective audio source enhancement system includes a processor and a memory, and a pre-processing unit configured to receive audio data including a target audio signal, and to perform sub-band domain decomposition of the audio data to generate buffered outputs. In addition, the system includes a target source detection unit configured to receive the buffered outputs, and to generate a target presence probability corresponding to the target audio signal, as well as a spatial filter estimation unit configured to receive the target presence probability, and to transform frames buffered in each sub-band into a higher resolution frequency-domain. The system also includes a spectral filtering unit configured to retrieve a multichannel image of the target audio signal and noise signals associated with the target audio signal, and an audio synthesis unit configured to extract an enhanced mono signal corresponding to the target audio signal from the multichannel image.

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 ofWaveNet 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 traditionalTTS systems, wherein each component requires laborious feature engineering and extensive domain expertise. Inference with system embodiments may be performed faster than real time.

The invention discloses a video real-time splicing device and method based on real-time conversation semantic analysis. The method includes the steps that a video material library and an audio material library which are related to standard semantics are established, a semantic analysis module conducts semantic analysis on input, a video unit and an audio unit which need to be selected for use are calculated according to the semantic analysis, and corresponding commands are generated and sent to a real-time splicing module for corresponding processing; the real-time splicing module calls the corresponding video unit and the corresponding audio unit from the video material library and the audio material library according to the commands of the semantic analysis module, picture splicing and audio synthesis are conducted according to the requirements of the commands, and complete audio and video fragments are formed and provided for a player so as to be played. By means of the device and method, videos and audio are spliced in real time according to the semantics, real-time interaction with visitors on the scene can be achieved, the content needing to be known by the visitors on the scene can be shown in the played audio and videos duly, and therefore the advertising effect is greatly improved.

The present invention relates to systems and methods for visual image audio composition. In particular, the present invention provides systems and methods for audio composition from a diversity of visual images and user determined sound data-base sources.

Disclosed are a communication server, a communication terminal and a voice communication method. The communication server includes an acquisition unit adapted to acquire audio output data respectively corresponding to different communication tools based on an audio hardware abstraction layer; a synthesis unit adapted to audio synthesis processing of the audio output data corresponding to the different communication tools so as to acquire audio synthesis data; and an output unit adapted to send the synthesis audio data respectively to the different communication tools. Multi-voice data combined and forwarded by use of the communication server and therefore interconnection and internetworking between communication terminals with different communication tools are achieved as well as multi-party calls.

The invention discloses a voice recognition-based network game online interaction system, which comprises a game server side and a plurality of game clients connected with the game server side, wherein the game server side comprises a voice acquisition module, a text input module, a channel selection module, a voice playing module, a text display module and a local cache module; and each game client comprises a channel management module, a control module, a voice recognition module and an audio synthesis module. According to the network game online interaction system, a player inputs voice or text through the game clients, transcribes the voice or the text into the text or the voice through the game server side, and outputs the text or the voice to the game clients of other players in the same chat channel, the other players in the same chat channel can receive the voice or text information at the same time no matter that the voice or the text is adopted for inputting, and each player can select the voice or the text to exchange according to the requirements, so that the voice recognition-based network game online interaction system is convenient to communicate and more humanized in design.

Method and system for audio synthesis of a digital data file representing an assembly of information-bearing sounds in digital form. One or more spaced apart data segments are designated as key blocks and are removed from the original data file. The remainder of the data file is encrypted or otherwise encoded and communicated to a selected recipient on a first channel. Locations, sizes and separation distances of key blocks from each other within the original data file and a selected portion of the encoding or encryption key are placed in a data supplement. The removed segments and data supplement (optional) are communicated to the selected recipient on a second channel and/or at another time. The original data file is recovered by using the data supplement information, or using already available information, decoding or decrypting the encoded or encrypted data file and replacing the removed segments within the data file remainder. Neither the remainder data file nor the removed segments plus data supplement is sufficient, by itself, to allow reproduction of the original data file. Each of the remainder data file and the removed segments plus data supplement can be distributed separately and subsequently combined when authorization or license to reproduce the sounds has been obtained.

The invention provides an audio synthesizer capable of converting voices to songs, which comprises a voice/audio recording mechanism for recording voices input by a user, digitally converting a preset voice, a downloaded voice or the input voices and storing as a voice sample; an audio selection mechanism for selecting audio frequencies from preset audio frequencies, audio frequencies recorded from outside and audio frequencies in downloaded audio frequencies, digitally converting and storing as an audio object; an audio synthesis mechanism for cutting the audio object into a plurality of syllables by taking the time length of the voice sample as one syllable and synthesizing the syllables of the voice sample and the audio object into synthesized audio frequencies; and an output mechanism for outputting a plurality of the synthesized audio frequencies.

The invention provides a cloud game live broadcast method and device. The cloud game live broadcast method comprises the following steps: generating a video for recording the operation process of a cloud game application program APP; receiving audio from terminal equipment, wherein the audio is used for recording voice information when a user operates the cloud game APP; obtaining an audio and video synthesized by the video and the audio; and encoding the audio and video and sending the encoded audio and video to a server of a live broadcast platform. According to the invention, clear game picture live broadcast and noise-free sound reproduction can be realized, the power consumption of the terminal equipment can be reduced, and a lag-free game service is provided.

A digital work of literature is vocalized using enhanced text-to-speech (TTS) controls by analyzing a digital work of literature using natural language processing to identify speaking character voice characteristics associated with context of each quote as extracted from the first work of literature; converting the character voice characteristics to audio metadata to control text-to-speech audio synthesis for each quote; transforming the audio metadata into text-to-speech engine commands, each quote being associated with audio synthesis control parameters for the TTS in the context of each the quotes in the work of literature;

and inputting the commands to a text-to-speech engine to cause vocalization of the work of literature according to the words of each quote, character voice characteristics of corresponding to each quote, and context corresponding to each quote.

The invention discloses an audio synthesizing device and an audio synthesizing method applied to the same. The audio synthesizing method comprises the steps as follows: obtaining first audio data of a song sung by a user and second audio data of the original; obtaining a tune difference according to the first audio data and the second audio data; judging whether the tune difference is larger than a threshold: synthesizing the first audio data and the second audio data if the tune difference is larger. With the adoption of the manner, the out-of-tune part of the user during singing can be covered up, and the experience effect of the user is improved.

The invention relates to an audio processing method and apparatus. The audio processing method comprises the following steps: acquiring at least two original audio files for communicating with contacts in an application; acquiring an audio synthesis instruction; and responding to the audio synthesis instruction, and synthesizing the acquired at least two original audio files into a target audio file according to a preset rule. A plurality of audio files in an instant messaging application are acquired, these audio files are synthesized into the target audio file, and after the target audio file is synthesized, effective voice messages are more concentrated, thereby being convenient for the management and storage of a user.