WO1996003829A1

WO1996003829A1 - Process and device for transmitting audio signals between communication terminals

Info

Publication number: WO1996003829A1
Application number: PCT/DE1995/000973
Authority: WO
Inventors: Istvan Sebestyen
Original assignee: Siemens Aktiengesellschaft
Priority date: 1994-07-25
Filing date: 1995-07-25
Publication date: 1996-02-08
Also published as: EP0772933A1; CN1157679A; KR970705278A

Abstract

A process for transmitting audio signals from one communication terminal (SCT) to another (RCT) through a communication network by using an appropriate packet-switched transmission protocol (HDLC LAP or packet media multiplexing) is characterised by the following steps: (a) a digital input audio signal is compressed in the sending communication terminal (SCT); (b) the compressed audio signal is subdivided into data blocks; (c) each data block is characterised as audio information data block or control data block; (d) audio information data blocks are transmitted by the sending communication terminal (SCT) to the receiving communication terminal (RCT) and stored therein by using an error-recognition, if required protected transmission protocol; (e) the stored audio information data blocks are outputted as digital audio signals or further processed by using a decompression process; (f) the control data blocks are exchanged between sender and receiver by said error-recognition, if required protected transmission protocol in order to control the communication procedure. Also disclosed are devices for carrying out said process with use-specific accessories.

Description

description

Method and arrangement for transmitting audio signals between communication terminals

The invention relates to a method for transmitting audio signals from a communication terminal via an optionally analog or digital communication network to another communication terminal (KEE) using a sighted transmission protocol as part of the ITU-T V.8 / V .δbis initialization and the ITU-T V.34 modem.

The via analog interfaces to an analog communication network, e.g. a telephone network or to a digital communication network, e.g. a mobile network, connectable telephone devices are mainly used for the transmission of analog audio / voice signals in quasi real time. After a connection has been established between two telephone terminals, the original acoustic signals are converted by a microphone of the transmitting device into analog electrical vibrations, which are transmitted to the receiving telephone terminal in an analog and simultaneous manner via the analog telephone network. In the case of digital communication, for example communication via a modem, the analog signals must also be converted into digital signals before transmission and converted back into analog signals after transmission. In the receiving telephone terminal, the received electrical audio / voice signals in the telephone receiver are likewise converted back at the same time into acoustic signals which contain audio / voice information similar to the original signals.

The object of the invention is to provide a safe, easy-to-implement method for audio signal transmission via a communication network which allows the connection to be established or routed via both analog and digital network components of the respective communication network

REPLACEMENT LEAF and enables the implementation of additional communication services.

In addition, an arrangement for carrying out such a method must be specified.

Advantageous effects of the invention are, in particular, a shortening of the transmission duration in real time relative to the duration of the audio information to be transmitted or an increase in the transmitted useful bandwidth even when transmitted in analog communication networks.

This object is achieved by a method having the features of claim 1. Favorable configurations are the subject of subclaims. In addition, an arrangement for

Implementation of special configurations of the method according to the invention specified in claim 8.

In the sending audio communication terminal, the audio / voice signals are converted into digital code. First, they are compressed in order to reduce the amount of information to be transmitted, and then they are divided into information data blocks and transmitted block by block - with the aid of corresponding protocols - without errors via the telecommunications network. The HDLC transmission method ("High Level Data

Link Control ") HDLC LAP (" Link Access Protocol ") - variants specified in the ITU-T series G, Q, T, V, X - and the packet media multiplexing method of the ITU-T series H are characterized by the definition of virtual channels that provide different types of information independently

"virtually continuously" divided into blocks - either secured or unsecured - transmitted.

The received information is decompressed in the received audio communication terminal and decoded into the original digital code, which results in digital audio / voice information that is identical to the original. From the receiving end

ER _SA TZBLATT The received information is usually temporarily stored on local digital storage devices, such as on a magnetic disk, in order to be output contiguously if necessary, for example after digital-analog implementation.

When using the method according to the invention, bidirectional information exchange is possible in principle in both directions. In half-duplex mode, audio / voice information can be exchanged alternately between the terminals either in the same connection or after each connection has been re-established. In full duplex mode, audio / voice information can be exchanged between the terminals at the same time.

The following special features of the speech compression standard enable particularly advantageous forms of implementation of the invention:

The current standardization of speech encoders with very low bit rates e.g. for picture telephony in the public telephone network in ITU-T leads to high quality speech coders with transmission speeds of up to 4-8 kbit / ε, which is preferably a secure transmission type, e.g. require according to HDLC LAP, but almost achieve the quality that can be achieved with digital signal transmission according to the CCITT G.726 recommendation.

The audio compression methods of audio signals of higher quality currently standardized according to ITU-TSG 15 and ISO / IEC JTC1 SC29, for example up to 16 kHz HIFI audio with transmission speeds of 24-32 kbit / s with a secured transmission type.

A method according to the invention enables, for example, the simple implementation of “voice memory telephones”, which can also be used when using analog subscriber network interfaces, i.e. Telephone devices with voice mail applications

REPLACEMENT LEAF fertilizer using suitable communication protocols.

In this application, in addition to ordinary telephony, a voice or sound message can be transmitted from the transmitter to the receiver or, in the case of an "audio / voice broadcast", to several receivers. Depending on the tone / speech compression scheme used, voice and / or tone messages with different tone quality can be transmitted.

With compression methods known today, a compression with respect to the time base by a factor of 4 can be achieved. When using modems with a high transmission rate of, for example, 28-30 kbit / s, real-time audio information of, for example, 7 kbit / s can consequently be used in one order using suitable protocols via the public switched telephone network (PSTN) the transmission time is reduced by a factor of 4 compared to the real-time transmission. For example, the actual transmission duration of a voice message, which would take four minutes under normal call conditions, could be reduced to one minute. The highly compressed telephone voice messages are also well suited for compact digital storage. In the example above, one minute of speech would take up approximately 52 kbytes of storage space in a "speech storage server". A special data compression / decompression circuit for the speech signal memory can thus be saved.

Another advantage of the invention is that, for example, with the aid of the sound compression methods developed by ITU-TSG 15 and ISO / IEC JTC1 SC29, audio signals of a higher quality, such as e.g. 7 kHz or 16 kHz audio can even be transmitted simultaneously using the method according to the invention via the analog telephone network with a useful band range from 0.3 to 3.4 kHz.

ER _S ATZBLATT The invention is based on particularly favorable

Implementations of the invention with reference to the

Figures explained in more detail:

It shows:

1 shows the block diagram of an audio communication terminal KES, KEE for realizing a "voice mail" connection (hereinafter referred to as "voice memory telephone");

2 shows the block diagram of an audio communication terminal KES, KEE according to FIG. 1, additionally with the possibility of accessing a voice mail server; FIG. 3 different types of connection between voice storage telephones KES, KEE, namely directly between two voice storage telephones KES1, KEE1 and between two voice storage telephones KES2, KEE2 via a voice storage server VMS, each including a communication network KN;

Figure 4 illustrates the V.8 "Start up" procedure of session initialization. This is absolutely necessary in the context of a V.34 modem. The V.8 "Start up" procedure is activated once, at the beginning of the communication connection.

FIG. 5 shows table 3 / V.8 of the "call function" category of an embodiment of the extended ITU-T V.8 initialization procedure;

FIG. 6 shows the table 4 / V.8 of the "modulation modes" category of the extended ITU-T V.8 initialization procedure;

FIG. 7 shows the table 5 / V.8 of the coding of the protocol category of the extended ITU-T V.8 initialization procedure;

During the connection, the "voice mail phone" is either with another "voice mail phone" terminal (KEE1 to KESl) or with a "voice mail server" (FIG

REPLACEMENT LEAF 3) (KEE2 to VMS, or KES2 to VMS) connected. The connection KEE1 to KES1 represents a direct connection of the "voice mail telephones", the information being exchanged without a time delay. The connection KEE2 to VMS represents a direct connection of the "voice memory telephone" (KEE2) to the "voice memory server" (VMS), the information being sent from the KEE2 to the VMS without a time delay. In contrast, the "voice memory telephone" KES2 accesses the VMS at any later point in time, in which case the information is sent from the VMS to KES2 without further time delay.

FIG. 1 contains a simplified block diagram of such an audio communication terminal: KES (transmitter) and KEE (receiver) have the same structure. The network interface is used for the physical connection of KES or KEE to the communication network (KN). The system control enables the switch from normal telephone to voice memory telephone mode and back. It also contains those elements of the system that are also included in a classic telephone terminal (e.g. dialing, ringing, switching on the microphone / loudspeaker (handset) when lifting, switching off when hanging up). Switch (S1) switches the microphone from the normal telephone to the "voice mail telephone" (VMP). The voice memory telephone mode is initiated with the help of ITU-T V.8bis / V.8. The audio / speech encoder (e.g. ITU-T G.723, G.8kbit / s) converts the analog voice signals into digital codes at the transmitter (KES), which is compressed. The compressed signals are prepared for transmission in the multiplexer, with not only the voice but also control information being exchanged. In the block marked with HDLC LAP X, the multiplexed information is transmitted securely via an ITU-T V.34 modem. At the receiver (KEE), the process takes place in reverse order. The HDLC-LAP-X block ensures the correct reception of the multiplexed audio / control information. The multiplex block demultiplexes the signals in control and voice files. The voice information

REPLACEMENT LEAF tions are decoded in the audio / speech codec and output via the switch (S2) on the loudspeaker / handset.

FIG. 2 shows the simplified block diagram of one “voice memory telephone” communicating with a “voice memory server”: the description of FIG. 2 is a superset of FIG. 1. A “voice memory telephone server” is on the communication network (KN) ( VMS) connected. An additional switch (S3) in the transmitter (KES) enables the entry of a "mailbox address" (subaddress) using the telephone keypad.

This address is passed on to the multiplexer in order to convert it accordingly into control information and to integrate it into the multiplexed data stream. With the help of this information, the "voice mail telephone message" is stored in KEE's "mail box" in the VMS. When listening, KEE uses the switch S3 to indicate its "mail box" number. This is converted into corresponding control information by the multiplexer and used in the data stream. VMS then forwards the saved message to KEE. The output of the messages takes place as in FIG. 1.

The elements of telecommunications standardized or to be standardized in the following are preferably used:

• Modem ITU-T V.34 with a higher transmission speed (up to 28000-33000 bit / s), and the associated initialization procedure according to ITU-T V.8 / V.δbis.

• An extension of the ITU-T V.8 / V.8bis recommendations necessary for the listed applications.

• The (optional) error-free transmission of information blocks (an HDLC-LAP-compliant, packet-switched transmission protocol ITU-T V.gmax or packet media multiplexing ITU-T H.223).

REPLACEMENT LEAF • The current standardization of speech coders (ITU-T G.723, G.dsvd, G.729) with a very low bit rate for picture telephony in the public switched telephone network in ITU-T has led to speech coders (quality approximately CCITT G.726) with a transmission rate of 5 - 10 kBit / s, which require a transmission type that is as secure as possible. The future 4 kbit / s speech coder will e.g. Currently also standardized by ITU-T SG15 and ISO / IEC JTC1 SC29.

• The current standardization of high quality audio encoders (up to 16 kHz audio bandwidth) with the bit rate 24-32 kbit / s in the public telephone network in the ITU-T and in the ISO / IEC, which also requires an error-proof transmission type .

To implement the present invention, modules of the above-mentioned standards are removed, combined, expanded and put together to form the overall system.

Some properties of communication terminals to be provided to implement a method according to the invention are explained in more detail below.

To implement a voice mail service using a method according to the invention, a telephone device with "voice memory" capability, a so-called "voice memory telephone", is required.

In the application mentioned, in addition to ordinary telephony, a voice or audio message in one

"Terminal-to-terminal" methods are transmitted from the sender to the receiver or in a "voice-message broadcasting" method (Voice Mail Broadcast) to several recipients. Depending on the audio / speech compression scheme used, speech and / or audio messages with different sound quality can be transmitted.

REPLACEMENT LEAF A typical terminal for this contains at least the elements of a telephone for the public switched telephone network or a radio telephone and additionally a communication control device for controlling the data transmission using an HDLC LAP transmission protocol to ensure error-free communication, an ITU-TV.34 modem and a device for carrying out a suitable ITU-T V.8 /V.δbis initialization procedure. At least the sending communication terminal (KES) also contains an audio coding device for compressing the audio input digital signal and at least the receiving communication terminal (KEE) also contains an audio decoding device for decompressing the received audio information data. Figure 1 contains a simplified block diagram of such an audio communication terminal. During the connection, the “voice mail phone” is either connected to another “voice mail phone” terminal or to a “voice mail server” (FIG. 3). FIG. 2 shows the simplified block diagram 1 with a “voice memory telephone” communicating with a “voice memory server”.

The audio message to be sent is processed in the sending audio communication terminal KES for communication between two “voice memory telephones”.

The preparation by processing the "voice memory telephone" message is carried out "off-line", ie before the connection is established. The sending subscriber switches his audio communication terminal KES to "voice mail phone mode", lifts the handset and speaks the message on the phone. The message is compressed and stored locally in the sending audio communication terminal KES. In a favorable embodiment, eavesdropping and local editing are possible. The message length is determined by the phone's internal memory. At least 256 kbytes of memory capacity (approx. 5 minutes of speech) should be provided. After the message has been processed, the actual communication can begin immediately or - for example to use the lower night tariffs - with a pre-programmed delay.

The connection is established as a connection between the "voice memory telephones". That means the sending

"Voice mail phone" first tries - without delay - to address the called device with signaling according to ITU-T V.8 /V.βbis as the receiving "voice mail phone". Extensions in ITU-T V.8 / V.δbis must be provided for this. An exemplary embodiment of an expanded ITU-T V.8 / V.8bis initialization is described below. If the called terminal answers as a "voice mail phone" (also with the help of ITU-T V.8 /V.8bis), the digital connection for information exchange can be established with the help of ITU-T V.34. If the called device is a "normal telephone", e.g. is a fax machine, the connection is terminated after a predetermined period of time. Possibly. a subscriber triggering a "voice memory telephone" call - until the time has expired - after hearing the end of the signaling tones (according to ITU-T V.8 /V.δbis), switch to the normal "telephone mode" manually, if he notices that the called telephone has been operated manually. In this case the connection will continue as a normal "phone call".

If two "voice mail phones" meet, then according to ITU-T V.8 / V.δbis the end device properties are exchanged between the sender and receiver in order to signal that "voice mail phone" communication is possible. It is also determined whether a half or / and full duplex ITU-T-V.34 connection can be supported.

After successful ITU-T V.β- / V.8bis signaling, the ITU-T V.34 procedure of the modems is initiated. After "line probing" and "training" in full or half duplex mode (i.e. determining the line quality and testing the maximum

REPLACEMENT LEAF suitable modem speed), the highest modem speed is also selected according to the rules of ITU-T V.34. If the half-duplex mode is selected, the V.34 parameters are then exchanged again according to ITU-T V.34 and the control commands required for communication are exchanged at 1200 bit / s. For the full duplex mode, the control commands must be exchanged as part of the data transfer. For this purpose, a "supervisory control" control channel must be defined. The data packets of this channel are marked as (CON). The following are considered at least as tax commands:

• Exchange of additional device properties of the sender and the receiver (e.g. labeling of audio / voice compression, mailbox capability, half / full duplex capability, notification of reception preferences)

• Setting the common, preferred transmission or reception parameters on the basis of the declared device properties of the transmitter and receiver, opening and closing of the data user channels

Send a message

Get a message

Addressing a voice mail box.

Normal termination of communication • Communication aborted.

The actual communication between sender and receiver corresponds to the data phase of ITU-T V.34 after the successful "ITU-T V.34 Resynch". Here, communication is carried out with the previously determined maximum speed. For the

"Voice mail phone" communication requires an error detection and possibly correction mode. This can be achieved with an HDLC LAP (G.vmax), or with packet media multipleing according to ITU-T H.223. To do this, the digitized and compressed voice information must be stored in blocks with, for example, 200 bytes per block and numbered

REPLACEMENT LEAF be rated according to HDLC LAP. If a block was identified as defective, it can be retransmitted.

In the analog telephone network, the transmission of a given voice mail message can take about a quarter of the time of an ordinary telephone connection, depending on the actual transmission speed and the need for error correction. During reception, the reception of a "voice mail phone" should indicate the receipt of a voice mail message.

After receipt of the message, that is, when all blocks have arrived at the receiver, the connection is cleared with the aid of a control command in accordance with the end of the data phase of ITUT-V.34. A display can be provided in the receiving "voice mail phone" to signal that there is a received voice mail message. Thus, the receiving subscriber can pick up his handset at any time, press a button and the message is decompressed and converted "back-line", ie not connected to the telecommunications network, into audible voice signals. The received message, possibly for storage and further processing, can be transmitted via an optional external interface on the "voice memory telephone", for example according to ITU-T V.24, with a connection to a personal computer. Likewise, "voice mail telephones" can possibly connect a data terminal device to the communication network with such or a similar interface. No separate modems would be required for the telephone terminal and the data terminal equipment.

A further embodiment of the invention provides for communication between "voice mail telephones" and a "voice mail phone" server.

Here, the preparation of the "voice mail phone" message is basically carried out as described above

REPLACEMENT LEAF In addition, however, the recipient's "mail box address" is also entered, for example with the number keypad of the telephone. In half duplex mode, this is communicated to the receiver in 1200 bit / s "handshaking" after the V.34 parameter exchange, in full duplex mode in normal V.34 data transfer. Commands.

After the message has been prepared, a connection to the "voice mail server" can be established immediately. For example, to use lower nightly tariffs, a connection can also be initiated with a pre-programmed delay.

The connection is established as a connection according to ITUT-V.8 / V.8bis. The "voice mail server" registers itself in the ITU-T V.δ / V.δbis also as a "voice mail phone", whereby the exchange of the end device properties of the sender and receiver shows whether "voice mail phone" Communication between the sending "voice mail phone" and the receiving one

"Voice mail server" is possible. If so, the ITU-T V.34 "handshaking" procedure ("line probing", "phasing", "training") is also used to select the highest possible modem speed according to the rules of ITU-T V.34 leads.

The actual connection between transmitter and receiver corresponds to the data mode of ITU-T V.34. An HDLC-LAP-compliant or packet media multiplexing protocol is required to ensure the best possible transmission. Under

Using the entered subaddress ("Subaddress"), the voice mail message is delivered to the "voicebox" of the receiving "voice mail phone".

After correct receipt of the message, the connection between the sending "voice mail server" and the forwarding "voice mail server" is disconnected.

E _RSA TZBLATT At a later point in time, the receiving “voice memory telephone” can use the corresponding “selective call” protocol element of the “voice memory telephone” to receive the voice memories received from the “voice memory server” (according to ITU-T V.δ / V.δbis) - Retrieve phone messages. For this purpose, a "voice mail telephone" with polling capability is required.

After a message has been called up from the "voice memory server", for example, a continuously lit red lamp can indicate that this received voice memory message is ready for use. The receiving party can then cause the message to be decompressed and converted back into audible speech signals. Here, the receiving communication terminal works “off-line”. The received message can in turn be used for further storage via optionally external interfaces (such as ITU-T V.24, RS-232) on the receiving "voice mail telephone" with a connection to a data terminal device, such as a personal computer ¬ security / processing are transferred.

Furthermore, an exemplary embodiment for the ITU-T V.δ (Appendix 1) and ITU-T V.34 parameterization is proposed with the addition of extensions. The V.8bis initialization procedure is similar (Appendix 2):

Appendix 1 and Appendix 2 are part of the description.

Figure 4 illustrates the V. δ "Start up" procedure of session initialization. This is absolutely necessary in the context of a V.34 modem. The V.8 "start-up" procedure is activated once at the beginning of the communication connection. As can be seen from FIG. 2, in a connection according to V.δ (and then V.34) after connection connection, the delay on the calling end device (DCE) side is approximately 1 second. The Cl signals (CI = "call identicator") are then transmitted at 300 b / s until the called terminal (DCE) does not respond with an ANSam signal ("user tone signal" from V.δ). Then the calling DCE sends the signals CMs ("Call Menu" signals, with the selection of the so-called "Call Function" categories - eg "voice mail phone", with the selection of modem modulation - eg for V.34 full duplex, with the selection of the protocol category - LAP "?"). The called DCE reflects - with its "JM" signals - the same type of information in order to declare its device properties. The two tables are used to determine whether communication between the two partners is possible. If so, the V.34 procedure usually follows within 2 seconds.

If the called DCE is not a "voice mail phone", but e.g. a normal telephone, the V.δ procedure is terminated after a certain period of time (timer). This time period is not specified in the V.β, but in the application of the "voice mail telephone".

In the current version of ITU-T V.δ, the CM / JM code assignments of the "voice mail telephone are missing. These assignments are made on request by the ITU. Below is an example of how to make them could.

Table 3 / V.δ (see Fig. 5) specifies the "Call

Function "category of V.δ. In the first octet are from the

REPLACEMENT LEAF ITU already reserved some code points. For the

"Voice mail phone" two code points are to be reserved. The code points shown were used as examples.

The modulation modes are specified in TABLE 4 / V.8 (see Fig. 6).

The octet for possible "protocols" can also be omitted if only one protocol is specified for the "voice mail phone" service. If several protocols are used, eg for different network types (PSTN, mobile), then uses this octet and specifies corresponding code points for the protocol types.

The GSTN access category is defined in TABLE 6 / V.δ (not shown). It indicates whether the calling or the called terminal is connected via the mobile telephone network.

The exchange of additional information is not intended by ITU-T V.δ, since V.8 is too slow for the exchange of larger information (300 b / s). For this reason, all additional negotiations or control commands should run via the procedures of the ITU-T V.34 modem.

After the V.8 "start-up" has run successfully, a V.34 "start-up" also runs. This contains:

V.34 "Line Probing" (line quality estimation) - V.34 HDX (half-duplex) or FDX (full-duplex) "Training" V.34 parameter exchange

REPLACEMENT LEAF V.34 resynchronization.

Only then does V.34 switch to its data transfer mode, where the actual user data is exchanged.

A so-called "retrain" from V.34 takes place if the line quality should not be sufficient during data communication. The data transfer is interrupted and a new transmission speed of sufficient quality is set via "Retrain". Then the transfer of the data continues. In the current version of V.34, the user data (e.g. audio information) cannot be transmitted simultaneously during the "retrain".

After successful ITU-T-V.8 signaling, the ITU-T-V.34 procedure of the modems is initiated. After "line probing" and "training in full or half duplex mode" (ie determining the line quality and testing the maximum suitable modem speed), the selection of the highest modem speed is also carried out according to ITU-T V. rules. 34 performed. Afterwards (only for half-duplex mode) the V.34 parameters are exchanged again according to ITU-T V.34 and the control commands required for communication are exchanged at 1200 bit / s. For full duplex mode, the control commands must be exchanged as part of the data transfer.

To do this, a virtual "Supervisory Control" channel must be defined. The data packets of this channel are specially marked. The following are at least the following as control commands:

REPLACEMENT LEAF Exchange of additional device properties of the transmitter and the receiver, such as, for example, marking the audio / voice compression, mailbox capability, half / full duplex capability, notification of the reception preferences; • Setting the common, preferred transmission or reception parameters based on the declared device properties of the transmitter and receiver; Opening and shooting of virtual user channels (eg for audio).

• sending a message; • retrieve a message;

• addressing a voice mail box;

• Normal termination of communication;

• Communication broken.

Replacement sheet

Claims

claims

1. Method for transmitting audio signals from a communication terminal (KES) via a communication network to another communication terminal (KEE) using a secured transmission protocol within the framework of ITU-T-V.δ- or ITU-TV. δbis initialization and the ITU-T V.34-MOD with the following procedural steps: a) a digital input audio signal is compressed in the sending communication terminal (KES); b) the compressed audio signal is divided into data blocks; c) each of these data blocks is identified as an audio information data block or a command data block; d) the audio information data blocks are transmitted from the sending communication terminal (KES) to a receiving communication terminal (KEE) using a protocol according to an error-detecting, if necessary, secured transmission method) and are stored in the latter; e) the stored audio information data blocks are output or processed as a digital audio signal after application of a decompression method, f) the command data blocks are securely exchanged between the transmitter and receiver using a protocol in accordance with the above-mentioned, error-recognizing, if necessary, secured transmission method between transmitter and receiver ¬ to control procedure.

2. The method according to claim 1, characterized by such a data compression that the transmission duration of the compressed audio signal, depending on the data block transmission rate between the communication terminals (KEA, KEE), is shorter than the duration of the digital input audio signal les.

REPLACEMENT LEAF

3. The method according to any one of the preceding claims, da¬ characterized in that the audio signal is a voice signal.

4. The method according to any one of the preceding claims, characterized in that the identification as an audio information block or command data module is carried out in a control field.

5. The method according to claim 4, characterized in that a label for audio information data blocks is provided and also a label for voice information data blocks and a label for the command data blocks.

6. The method according to any one of claims 4 or 5, characterized ge indicates that in a full duplex mode, the command data blocks intended for control together with the audio information data blocks from the sending communication terminal (KES) to the receiving communication terminal (KEE) are transmitted in the user data channel.

7. The method according to any one of the preceding claims, characterized by the use of the initialization process according to ITU-T V.δ / V.δbis and the use of an error-recognizing, if necessary, secured transmission protocol as part of a data transfer of an ITU-T V .34 Modems.

8. Arrangement for carrying out a method according to one of the preceding claims, consisting of a communication network (KN) and a transmitting communication terminal (KES) and a receiving communication terminal (KEE), each of which has at least the usual telephone equipment, characterized in that that the communication terminals (KEE, KES) also have a communication control device for controlling the data transmission using the ITU-T V.8 / V.8bis initialization process, an error-

REPLACEMENT LEAF Knowing, if necessary, secured transmission protocol and the ITU-T V.34 modem that at least the sending communication terminal (KES) also has an audio coding device for compressing the audio input digital signal and that at least the receiving communication terminal (KEE) has an audio Decoding device for decompressing the received audio information data.

Spare sheet