Abstract: A method that allows two or more APs (application programs) to use a sound card and an AP to select an input device of the sound card regardless of a mixer device of the sound card is disclosed. A connecting module 31 is disposed among a sound card 17, a HDD 13, and an NIC 19. Switches 32 and 33 are disposed in parallel to the connecting module 31. In other words, an audio signal that is input form the sound card 17 is supplied to the NIC 19 through the switch 32. In addition, the audio signal is supplied to the HDD 13 through the switch 33. The switch 32 is equivalent to a concept of an application interface for starting and stopping a sending operation for an audio signal supplied from the sound card 17 to the network 111. The switch 33 is equivalent to a concept of an application interface for starting and stopping a saving operation for an audio signal supplied from the sound card 17 to the HDD 13.

Abstract: A system and method is described for transcribing a recorded message automatically from a voicemail message system or an answering machine. The designated message may be automatically transcribed and forwarded to various archival devices, such as a printer, a facsimile machine, computer memory storage, or email. Thus, the system and method of the present invention automates the transcription process of a recorded message eliminating the need for a person to transcribe the message.

Abstract: A method for identifying a translation transforms a representation of context words in a source language into a transformed representation of context in a target language based on multiple different translations for at least one context word. Candidate representations of context words in a target language are determined for at least two candidate translations. The transformed representation is used to score each candidate representation and the scores are used to select a candidate translation. In another aspect of the invention, a method and apparatus are provided for updating a translation dictionary by receiving votes from multiple clients as to which translation is correct.

Abstract: The present invention relates to a transmitter and a receiver for speech coding and decoding by using an additional bit allocation method. The transmitter and the receiver according to the present invention realize a voice communication service of high quality by using additional bits permitted in system requirements while using a conventional speech coder as it is. In addition, the transmitter and the receiver according to the present invention have an advantage in that they enable insertion of additional quantization blocks while not changing the structure of the conventional standard speech coder, since they allocate additional bits by applying a multi-stage quantization procedure not in a speech signal domain but in a parameter domain.

Abstract: A system analyzes text, determines phrasing and, in an exemplary embodiment, reformats the text to establish optimal spacing and related features for readability, reader comprehension and publishing economies. A neural network uses a library of text data to analyze text and determine phrases. Formatting emphasizes phrases using one or more of a plurality of techniques including word spacing, text darkness and controlling line breaks.

Abstract: A method and apparatus determine a channel response for an alternative sensor using an alternative sensor signal, an air conduction microphone signal. The channel response and a prior probability distribution for clean speech values are then used to estimate a clean speech value.

Abstract: The present invention deals with noisy data not by eliminating low frequency dependency structures, but rather by weighting the dependency structures. The dependency structures are weighted to give less weight to dependency structures which are more likely incorrect and to give more weight to dependency structures which are more likely correct.

Abstract: A system and method for detecting speech utilizing audio and video inputs. In one aspect, the invention collects audio data generated from a microphone device. In another aspect, the invention collects video data and processes the data to determine a mouth location for a given speaker. The audio and video are inputted into a time-delay neural network that processes the data to determine which target is speaking. The neural network processing is based upon a correlation to detected mouth movement from the video data and audio sounds detected by the microphone.

Abstract: The invention relates to a method for transcoding audio signals in a communications system. In order to improve the inter-operability between units (2,40) capable of handling wideband audio signals and units (3,46) or network components (50) capable of handling narrowband audio signals, it is proposed that first, an audio signal is received in a network element (42) of a communications network via which said audio signal is transmitted. Next, it is determined in said network element (42) whether a transcoding of the received audio signal is required. In case a narrowband-to-wideband transcoding of the received signal is required, the received narrowband audio signal is transcoded into a wideband audio signal in the network element (1,42). The generated wideband audio signal is then forwarded to the receiving terminal (2,40). The invention equally relates to a corresponding communications system and its components.

Abstract: A mapping transform unit subjects input audio signals to a mapping transform and generates frequency region signals that take frequency as a variable; a code amount designation unit supplies a preset coding bit rate as a code amount output; a frequency region signal compression encoder, based on the code amount, subjects input frequency region signals to a compression encoding process and generates a bitstream; and a bandwidth-limiting unit executes a bandwidth-limiting processing in which a part of the frequency zone covered by frequency region signals is allotted to an attenuation frequency zone, and in which the value of the frequency region signal is multiplied by an attenuation coefficient having a value less than 1 in the attenuation frequency zone to attenuate the frequency region signal in the attenuation frequency zone, and supplies the frequency region signals that have undergone the bandwidth-limiting processing to the frequency region signal compression encoder.

Abstract: A statistical machine translation (MT) system may use a large monolingual corpus to improve the accuracy of translated phrases/sentences. The MT system may produce a alternative translations and use the large monolingual corpus to (re)rank the alternative translations.

Abstract: According to the present invention, apparatus and methods are provided to enable a user to locate and retrieve sets of information relevant to search criteria specified in a search query submitted by the user. Search results include not only a list of information sets matching with the search criteria, but also the preserved structure of any tags used in annotating the information set according to a structured mark-up language such as XML. A user may select a tag from a presented list of the returned tag structures, and the apparatus lists those documents containing the selected tags. The list of tags is then adjusted to include the selected tag and any other of the returned tags contained in the listed documents. Further tag selection from the adjusted list leads to a further refinement of the listed documents, enabling the user to navigate the search results on the basis of tag information.

Abstract: At the coder side, bits of samples of each frame of an input digital signal are concatenated every digit common to the samples across each frame to generate equi-order bit sequences, which are output as packets. At the decoding side, the input equi-order sequences are arranged inversely to their arrangement at the coder side to reconstruct sample sequences. When a packet dropout occurs, a missing information compensating part 430 correct the reconstructed sample sequences in a manner to reduce an error between the spectral envelope of the reconstructed sample sequence concerned and a known spectral envelope.

Abstract: A multiple step adaptive method for time scaling. Synthesizing S3[n] signal from signal S1[n]signal and S2[n]signal. Comprising following steps: (a) calculating a first magnitude of a cross-correlation function of S1[n]signal and S2[n]signal according to a first index; (b) comparing the first magnitude with a threshold value; (c) if first magnitude is smaller than threshold value, calculating a first reference magnitude of cross-correlation function of S1[n]signal and S2[n]signal according to a first reference index behind the first index by a first determined number, or calculating a second reference magnitude of the cross-correlation function of the S1[n] signal and the S2[n] signal according to a second reference index behind the first index by a second number; (d) synthesizing the S3[n] signal by adding S1[n]signal to the S2[n] signal in accordance with a maximum index corresponding to a largest magnitude among all the magnitudes calculated in (c).

Abstract: A codebook excited linear prediction coding system providing improved digital speech coding for high quality speech at low bit rates with side-by-side codebooks for segments of the modeled input signal to reduce the complexity of the codebook search. A linear predictive filter responsive to an input signal desired to be modeled is used for identifying a basis vector from a first codebook over predetermined intervals as a subset of the input signal. A long term predictor and a vector quantizer provide synthetic excitation of modeled waveform signal components corresponding to the input signal desired to be modeled from side-by-side codebooks by providing codevectors with concatenated signals identified from the basis vector over the predetermined intervals with respect to the side-by-side codebooks. Once a codevector is identified, the codebook at the next segment is searched and a concatenation of codevectors is provided by selecting up to but not including the current segment.

Abstract: The present invention provides an apparatus, method and tangible medium storing instructions for determining tonality of an input audio signal, for selection of corresponding masked thresholds for use in perceptual audio coding. In the various embodiments, the input audio signal is sampled and transformed using a compressed spectral operation to form a compressed spectral representation, such as a cepstral representation. A peak magnitude and an average magnitude of the compressed spectral representation are determined. Depending upon the ratio of peak-to-average magnitudes, a masked threshold is selected having a corresponding degree of tonality, and is used to determine a plurality of quantization levels and a plurality of bit allocations to perceptually encode the input audio signal with a distortion spectrum beneath a level of just noticeable distortion (JND).

Abstract: Methods and apparatus are provided for the creation and utilization of unique compressed data stream compositions, structures and formats which allow for the alteration of the data stream's data rate without first decoding the data stream back to its uncompressed form and then re-encoding the resulting uncompressed data at a different data rate. Such methods and apparatus perform this data rate alteration, known as scaling, such that optimal quality is maintained at each scaled data rate, while performing said scaling with low computational complexity. In addition, the present invention provides for data rate alteration in small increments. A unique application for the disclosed bit rate scaling method and apparatus is also described.

Abstract: A speech processing apparatus able to enhance formants more naturally, wherein a speech analyzing unit analyzes an input speech signal to find LPCs and converts the LPCs to LSPs, a speech decoding unit calculates a distance between adjacent orders of the LSPs by an LSP analytical processing unit and calculates LSP adjusting amounts of larger values for LSPs of adjacent orders closer in distance by an LSP adjusting amount calculating unit, an LSP adjusting unit adjusts the LSPs based on the LSP adjusting amounts such that the LSPs of adjacent orders closer in distance become closer, an LSP-LPC converting unit converts the adjusted LSPs to LPCs, and an LPC combining unit uses the LPCs and sound source parameters to obtain formant-enhanced speech.

Abstract: Generating a transfer dictionary used in a transfer-based translation machine system. A pair of source/target language sentences are received. The source language sentence comprises at least one marked idiom, at least one argument and at least one marked collocation. The target language sentence comprises the target language translation for the idiom and the source language word(s) for the argument. The source language sentence is parsed to generate a source language syntactic tree. Nodes are extracted from the source language syntactic tree. A least common ancestor node of the extracted nodes is calculated and source language structure information is generated based on the source language syntactic tree data structure. Target language structure information is generated by adding the part-of-speech information to each morpheme in the target language sentence and by replacing each source language word in the target language with the corresponding syntactic information within the source language syntactic tree.

Abstract: A method (300) of editing program code is proposed. The program code consists of main instructions written in a high-level language. Typically, the program code further includes one or more sets of service instructions, which are not directly connected to a desired algorithm (such as tracing instructions for debugging purposes, performance measuring instructions, and the like). In the solution of the invention, each set of service instructions is enclosed (321-322) between a pair of predefined comments defining a starting tag and an ending tag. An editor of the program code is provided with commands for automatically disabling (341) the service instructions (converting them into corresponding comments); at the same time, the disabled instructions are condensed (340) on the monitor. Additional commands are available for automatically enabling (368) and restoring (367) the service instructions.