Abstract: A system for automatically merging multiple time-stamped transcriptions is provided. The system includes a transcription server for receiving a signal having time-stamp information, a splitter, a merging utility, and a text output. A method for automatic merging of multiple time-stamped transcriptions comprises the following steps: transferring a signal having timestamp information encoded therein to a splitter which yields a mixed audio output having resultant corresponding audio channels, transferring the mixed audio output to a transcriber server which thereby yields one or more text outputs, and the text outputs being merged by a merging utility with the timestamps included in the signal thereby providing a single text file.

Abstract: A system for automatically merging multiple time-stamped transcriptions is provided. The system includes a transcription server for receiving a signal having time-stamp information, a splitter, a merging utility, and a text output. A method for automatic merging of multiple time-stamped transcriptions comprises the following steps: transferring a signal having timestamp information encoded therein to a splitter which yields a mixed audio output having resultant corresponding audio channels, transferring the mixed audio output to a transcriber server which thereby yields one or more text outputs, and the text outputs being merged by a merging utility with the timestamps included in the signal thereby providing a single text file.

Abstract: A signaling and media protocol for multi-channel recording using a client and a server is provided comprising an IP data network, a recording application on a client device, a signaling server and a media server. A method for signaling and media protocol for multi-channel recording is provided having a message type, the message type comprising means for command/response sequences being command and acknowledgement performed in the steps of Start Recording/acknowledge start recording, Current Recording/acknowledge current recording, Stop Recording/acknowledge stop recording. The method further comprises the steps of command/response sequences selectably performed in any order, as desired, of: Received Recording/acknowledge received recording, Mark Recording/acknowledge mark recording, Insert Recording/acknowledge insert recording, Noisy Recording/acknowledge noisy recording, Dropped Recording/acknowledge dropped recording, followed by Delete Recording/acknowledge delete recording.

Abstract: A system for a high performance audio codec provides higher voice quality and higher recognition accuracy from an ASR engine at an increased data rate and computational power and embodiments include those having a CELP-based codec, an ASR engine, a text comparator, an encoder, a decoder, an LPC Computation and formant analysis module, a dual stage data rate determination module, a VQ of LSP coefficients module, a pitch synthesis and optimal pitch parameter search module, and an excitation codebook parameter search module. A method for high performance audio codec includes three stages and comprises the steps of having an ASR engine yield transcribed text from each of an uncompressed reference signal and a decompressed signal that has passed through an encoder and wherein the transcribed text is compared with original text to determine word error rates in an iterative process whereby both voice quality and recognition accuracy are optimized.

Abstract: This invention produces data packets that can vary in length and/or data compression ratio. First, an algorithm is employed to transform a data signal into fixed or variable length data packets at variable data compression ratios. If the algorithm produces fixed length data packets, the fixed length data packets are then converted to variable length data packets, which include only the valid data bytes of the fixed length data packets. Finally, the variable compression ratio, variable length data packets are provided with length codes at each end of each data packet to facilitate bi-directional searching and decompression. The transition from fixed to variable length data packets employs a buffer which stores the fixed length data packets until the fixed length data packets are converted to variable length data packets.

Abstract: A system for automatically merging multiple time-stamped transcriptions is provided. The system includes a transcription server for receiving a signal having time-stamp information, a splitter, a merging utility, and a text output. A method for automatic merging of multiple time-stamped transcriptions comprises the following steps: transferring a signal having timestamp information encoded therein to a splitter which yields a mixed audio output having resultant corresponding audio channels, transferring the mixed audio output to a transcriber server which thereby yields one or more text outputs, and the text outputs being merged by a merging utility with the timestamps included in the signal thereby providing a single text file.