France Telecom

France Telecom Sa

Orange

David Deleam

Dominique Massaloux

Balazs Kovesi

Patrick Boissonade

Yannick Mahieux

Jean-Philippe Thomas

Arnault Nagle

Grégoire Le Tourneur

Franck Bouteille

Monique Auvray

Aurélien Sollaud

Laurent Clarimon

Grégolre Le Tourneur

Jean-Claude Schmitt

Olivieir Calliger

Jean-Marc Le Solliec

A method of notifying on a downstream link () of a network element () a transmission defect affecting an audio signal coded on an upstream link () of said network element. The method comprises a step in which the network element () sends, on said downstream link, an audio coded notification signal containing a specific notification pattern. Application to the switchover from an asynchronous link to a synchronous link.

Method of notifying a transmission defect of an audio signal

A method of concealing transmission error in a digital audio signal, wherein a signal that has been decoded after transmission is received, the samples decoded while the transmitted data is valid are stored, at least one short-term prediction operator and one long-term prediction operator are estimated as a function of stored valid samples, and any missing or erroneous samples in the decoder signal are generated using the estimated operators. The energy of the synthesized signal that is thus generated is controlled by means of a gain that is computed and adapted sample by sample.

Transmission error concealment in audio signal

The invention relates to processing signal for improving a sound pick-up in telephone and/or videophone units. The inventive method consists in comparing () a signal transmitted by a unit microphone (Mic) with a threshold low value (THR_LOW) and in amplifying () said microphone-transmitted signal when it is less than the threshold low value. However, said steps depend on the test () carried out on a signal used for feeding the loud speaker (HP), in particular consisting in controlling said last signal (crete_HP) of a background noise order (BDF) in such a way that the background noise is not amplified by an acoustical coupling between the microphone (Mic) and the loud speaker (HP).

Sound pick-up method and device, in particular for handsfree telephone terminals

The invention relates, in particular, to a system for managing communication sessions between first terminals adapted to send audio and/or video and/or data streams during these sessions and second terminals accessible via an external data transmission network. The first terminals are connected to a stream concentrator itself connected to the external network. Since the streams sent by at least some of the first terminals are coded or intended to be coded by hierarchical encoders, the system includes means for processing said streams so that each of them is sent to the external network by the concentrator at a given hierarchy level and means for determining said hierarchy level as a function of parameters relating to the number and nature of the sessions set up between the first terminals and the second terminals.

Communication session management system and a stream concentrator including such a system

This device for processing audio signals for telecommunication equipment comprises automatic gain control means () acting on the level of the received audio signal (Sr) by amplifying and attenuating the signal as a function of the level of the received signal and echo cancelling means (). The automatic gain control means () comprise amplification means (-) and attenuation means (-) for separately amplifying and attenuating the received signal downstream and upstream of the echo cancelling means (), when considering the propagation direction of the received signal.

Device and method for processing audio signals and telecommunication set equipped with such a processing device

A method of concealing transmission error in a digital audio signal in which a signal that has been decoded after transmission is received, the samples decoded while the transmitted data is valid are stored, at least one short-term prediction operator and one long-term prediction operator are estimated as a function of stored valid samples, and any missing or erroneous samples in the decoder signal are generated using the operators estimated in this way, the method being characterized in that the energy of the synthesized signal as generated in this way is controlled by means of a gain that is computed and adapted sample by sample.

Transmission error concealment in an audio signal

The invention relates to a real time processing and management method for canceling out the echo between a loudspeaker (HP) and a microphone (M) of a computer terminal. Said method consists of establishing the loudspeaker signal or the microphone signal as a reference signal and synchronizing (A) the other signal with said reference signal, and (B) synchronizing the acquisition and audio reestablishment tasks. The delay that exists (re) between the other signal and the reference signal is measured (C) and the existing delay value (re) is validated (D) in order to control the delay applied to the reference signal starting from the current delay (re). The invention can be used for computer terminals which are equipped with multi-tasks operating systems.

Real time processing and management method for cancelling out the echo between a loudspeaker and a microphone of a computer terminal

A conference system of the type that comprises a bridge () to which a plurality of terminals (to) can be connected to participate in a conference, said bridge () being equipped to receive the audio and/or video and/or data flows of each terminal (to) and to retransmit them to the other terminals, wherein the flows transmitted by said terminals (to) are coded by means of hierarchical coders and wherein said bridge () is equipped with means (toto) for processing the audio/video/data flows coming from the terminals (j) so that they have a hierarchical level (q) in the flow intended for the terminal (i) and means () for determining said hierarchical level (q) based on decision-making criteria relating to the status of said conference system.

Videoconference system

The digital data packets received contain data describing successive time segments of a signal stream to be restored, for example audio. Elements obtained from the packets received are recorded in a buffer memory organized in first-in first-out mode. A burst of packets fulfilling a certain burst size condition is detected a priori, on the basis of a reception delay evaluated in response to the reception of a first packet of the burst. The content of the buffer memory is controlled by accounting for the detection of the burst fulfilling the size condition.

Growing community of inventors

Perros Guirec, France

David Deleam