U.S. Philips Corporation

Koninklijke Philips Corporation N.V.

U.S. Philips Corporaion

Friedhelm Wuppermann

Fransiscus M J De Bont

Fransiscus M De Bont

Franciscus M De Bont

Robert J Sluijter

Eric Kathmann

Volker Stahl

Method of automatic recognition of a spelled speech utterance

In a CELP coder a comparison between a target signal and a plurality of synthetic signals is made. The synthetic signal is derived by filtering a plurality of excitation sequences by a synthesis filter having parameters derived from the target signal. The excitation signal which results in a minimum error between the target signal and the synthetic signal is selected. The search for the best excitation signal requires a substantial computational complexity. To reduce the complexity a preselection of a small number of excitation sequences is made by selecting a small number of excitation sequences resembling the most a backward filtered target signal. With this small number of excitation sequences a full complexity search is made. Due to the reduced number of excitation sequences involved in the final selection the required computational complexity is reduced.

Reduced complexity signal transmission system

In a CELP coder a comparison between a target signal and a plurality of synthetic signals is made. The synthetic signal is derived by filtering a plurality of excitation sequences from a one dimensional codebook by a synthesis filter having parameters derived from the target signal. The excitation signal that results in a minimum error between the target signal and the synthetic signal is selected. In order to reduce the complexity of the search for the best excitation signal, the selection is done in two stages. First a preselection of a small number of excitation sequences is made by selecting only every L.sup.th codebook entry for preselecting a plurality of excitation sequences. Thereafter, with this small number of excitation sequences, a fill complexity search is made in which all excitation sequences surrounding the preselected ones are involved in the selection.

CELP coding with two-stage search over displaced segments of a one-dimensional codebook

In a CELP coder, a comparison between a target signal and a plurality of synthetic signals is made. The synthetic signal is derived by filtering a plurality of excitation sequences by a synthesis filter having parameters derived from the target signal. The excitation signal which results in a minimum error between the target signal and synthetic signal is selected. The search for the best excitation signal requires a substantial computational complexity. To reduce the complexity, a preselection of a small number of excitation sequences is made using a reduced complexity synthesis filter. With this small number of excitation sequences, a full complexity search is made. Due to the reduced number of excitation sequences involved in the final selection, the required computational complexity is reduced.

In a CELP coder, a comparison between a target signal and a plurality of synthetic signals is made. A synthetic signal is derived by filtering a plurality of excitation sequences by a synthesis filter having parameters derived from the target signal. The excitation signal, which results in a minimum error between the target signal and the synthetic signal, is selected. The search for the best excitation signal requires a substantial computational complexity. To reduce the complexity, a preselection of a small number of excitation sequences is made by selecting a small number of excitation sequences resembling the most backward filtered target signal. With this small number of excitation sequences a full complexity search is made. Due to the reduced number of excitation sequences involved in the final selection the required computational complexity is reduced.

Encoder using an excitation sequence and a residual excitation sequence

In a CELP coder a comparison between a target signal and a plurality of synthetic signals is made. The synthetic signal is derived by filtering a plurality of excitation sequences by a synthesis filter having parameters derived from the target signal. The excitation signal which results in a minimum error between the target signal and the synthetic signal is selected. In order to reduce the complexity of the search for the best excitation signal, thee selection is done in two stages. First a preselection of a small number of excitation sequences is made using a reduced complexity synthesis filter. Thereafter, with this small number of excitation sequences, a full complexity search is made. Due to the reduced number of excitation sequences involved in the final selection the required computational complexity is reduced.

In a transmission system for transmitting speech and music signals, an input signal is coded in a coder (11) by a time domain coder (4). The output signal of the time domain coder (4) is decoded by a time domain decoder (8) and the signal thus decoded is subtracted from the input signal by a subtracter circuit (10). To improve the coding quality, the difference signal is coded by a frequency domain coder (12) and the output signal of the time domain coder (4) and the frequency domain coder (12) are combined in a multiplexer (14) and transmitted to a receiver.

Transmission system utilizing different coding principles

In a transmission system a signal is encoded in an encoder (7) and the encoded signal is transmitted by a transmitter (2) via a transmission medium (4) to a receiver (6). In the encoder, analysis parameters of the input signal are determined by an analyzer (8) and quantized by a quantizer (14). The transmitter transmits quantization level numbers to the receiver (6), and in the receiver decoded analysis parameters are derived by interpolating level numbers of two subsequent sets of analysis parameters, and subsequently determining the corresponding analysis. By interpolating the level numbers instead of the analysis parameters themselves, a substantial amount of computational complexity is saved.

Signal transmission system in which level numbers representing

In a CELP coder a comparison between a target signal and a plurality of synthetic signals is made. The synthetic signal is derived by filtering a plurality of excitation sequences from a one dimensional codebook by a synthesis filter having parameters derived from the target signal. The excitation signal which results in a minimum error between the target signal and the synthetic signal is selected. In order to reduce the complexity of the search for the best excitation signal, the selection is done in two stages. First a preselection of a small number of excitation sequences is made by selecting only every L.sup.th codebook entry for preselecting a plurality of excitation sequences. Thereafter, with this small number of excitation sequences, a full complexity search is made in which all excitation sequences surrounding the preselected ones are involved in the selection.

CELP coding with two-stage search over displaced segments of a

A transmission system for encoded transmission of audio signals with frequency band compression. An input signal to be transmitted is divided into two spectral portions in a coder. One spectral portion is digitally encoded by a time-domain coder, and the other spectral portion is coded by a transform domain coder. The two digitally coded signals are then multiplexed for transmission as a composite digitally coded signal to a receiver. At the receiver the composite signal is demultiplexed and the two spectral portions thereof are respectively decoded by a time-domain decoder and a transformer-domain decoder. The decoded signals, covering the respective spectral portions, are then combined to reconstitute the originally transmitted audio signal.

Transmission system implementing different coding principles

An audio signal digital transmission system wherein the signal is sampled and successive segments each of N samples are digitally encoded. Encoding has heretofore been effected by vector quantization, whereby a memory stores a book of a large number of codewords corresponding to possible segments of successive signal samples, and the best matching stored segment is determined by calculation. The present invention bases encoding on comparison of the envelope of a signal segment with a limited number of reference envelopes. The best matching reference envelope is selected, and an identifying codeword therefore is transmitted along with related parameters. A receiver reproduces the signal from the reference envelope and the related parameters. Since identified envelopes of the signal segments are less variable and simpler to evaluate than the signal samples in the segments, a considerable reduction of memory capacity and calculation complexity is achieved.

Growing community of inventors

Eindhoven, Netherlands

Friedhelm Wuppermann