Abstract: Telecommunication systems including a terminal, a switch and on Internet/Intranet having a memory for storing information blocks at locations defined by addresses, can be made more user-friendly by introducing a detector for detecting speech-recognition/non-speech recognition related parts in control signals originating from the terminal and defining the addresses or n response signals sent from memory to terminal and including said information blocks. A processor is coupled to the detector for routing, in dependence on the detection, routing (parts of) the signals to a server for either conversion of (parts of) the signals or for information purposes.

Abstract: In telecommunication systems including a terminal and a speech recognizer for vocal commanding, an indication signal originating from the terminal is detected, e.g., a telephone number dialed or a key signal or a vocal signal, or originating from an application located in the terminal or elsewhere in the network and in dependence of the indication signal a capacity parameter for the vocal commanding is adjusted, whereby a flexible capacity parameter is created for adjusting the available bandwidth between terminal and speech recognizer and/or for adjusting a processor capacity of terminal and/or speech recognizer, e.g., a sampling rate or a noise reduction being deactivated, as a result of which name dialing, command and control, and dictation can be done with the highest efficiency.

Abstract: Distributed Speech Recognition (DSR) systems comprise terminals having a complex preprocessing unit and a network having a final processing unit. By leaving a Fast Fourier Transformation (FFT) together with a filtering function and a compression in said preprocessing unit by shifting both nonlinear transformation as well as a Discrete Cosine transformation (DCT) from said preprocessing unit into said final processing unit, where both follow decompression, said terminals are of a lower complexity. By using a MEL-filter (melody-filter) which preferably is adaptable, said low complex terminals are made more flexible.

Abstract: Known telecommunication systems comprising a terminal and a speech recognizer for vocal commanding offer one fixed capacity for vocal commanding. By detecting an indication signal originating from said terminal, like for example a telephone number dialled or a key signal or a vocal signal, or originating from an application located in said terminal or elsewhere in the network and by in dependence of said indication signal adjusting a capacity parameter for said vocal commanding, a flexible capacity parameter is created for adjusting the available bandwidth between terminal and speech recognizer and/or for adjusting a processor capacity of terminal and/or speech ercognizer, like a sampling rate or a noise reduction being (de)activated, as a result of which name dialling, command & control, and dictation can be done with the highest efficiency.

Abstract: Noise reduction systems comprising an input for receiving per time-interval input signals originating from a Fast Fourier Transformator (frequency-components+values/amplitudes) and comprising a noise estimator coupled to said input for performing noise estimations per input signal and comprising a converter coupled to said noise estimator for performing conversions of said noise estimations and for generating correction signals and comprising a combiner coupled to said converter and to said input for generating per time-interval output signals (input signals minus correction signals=frequency-components +values/amplitudes with reduced noise) do not take into account their surroundings and are rather static. By introducing adaptation signals for adapting said conversions, noise reduction systems become dynamic and more flexible.

Abstract: Telecommunication systems including a terminal for speech communication and a network with a speech recognizer may couple the terminal and the network via a packet-switched coupling, whereby the terminal is provided with a detector system for detecting voice activity in audio signals and with a processing unit for processing audio signals and with a buffer for buffering processed audio signals for generating further information blocks and with a network-combiner for combining the information blocks with the further information blocks for creating an information stream for the speech recognizer. Preferably, in the detector system, processed audio signals or threshold signals representing differences between processed audio signals or indication signals are used for forming information blocks.

Abstract: Known telecommunication systems comprising a network-device and a terminal coupled via a wireless link for exchanging information can be made more flexible and more individualized by storing a terminal-definition code in said network-device for defining terminal-characteristics, whereby at least a first part of said information should be made dependent upon said terminal-definition code. Preferably, a preferences code is also stored, whereby at least a second part of said information should be made dependent upon said preferences code. A third part of said information may comprise (distributed) speech (recognition) signals. Both codes could further be made adjustable by a user.

Abstract: Due to, after the introduction of distributed speech recognition, two kinds of terminals being in existence, novel terminals comprising a preprocessing unit for cooperation with a final processing unit, and old fashioned terminals not comprising such a preprocessing unit, switches will have to react differently in dependence of what kind of terminal needs to be switched. This can be avoided by locating said preprocessing unit between said old fashioned terminal and said final processing unit, for example between said old fashioned terminal and said switch or in said switch before the switching is done.