System for measuring and automatically compensating for the distorsions of a connection between a telephone apparatus and a central voice processing unit

• US 4,847,885 A
• Filed: 07/08/1988
• Issued: 07/11/1989
• Est. Priority Date: 07/10/1987
• Status: Expired due to Term

• Current Assignee: Ing. C. Olivetti & C., SpA (Telecom Italia S.p.A.)
• Original Assignee: Ing. C. Olivetti & C., SpA (Telecom Italia S.p.A.)
• Inventors: Vittorelli, Vittore
• Primary Examiner(s): SCHREYER, STAFFORD
• Application Number:
  US07/219,756
• Time in Patent Office: 368 Days
• Field of Search: 379/6, 379/16, 379/21, 379/22, 379/24, 379/27, 379/32, 379/89
• US Class Current: 379/22.02
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 3/30 for subscriber's lines , for the local loop
• CPC Class Codes: H04M 3/323 for the arrangements providing the connection (test connection, test call, call simulation)
• CPC Class Codes: H04M 3/40 Applications of speech amplifiers
• CPC Class Codes: H04M 3/533 Voice mail systems

Abstract

The central unit (14) comprises a distortion sampler (16) including a spectral sound analyzer and a memory for recording the digital data of at least one reference sound. The central unit by means of the analyzer analyzes a sequence of such sounds which are received by way of the telephone line (11, 13) and makes a comparison with the recorded data. It thus produces a measurement of the frequency response of the transmission, that is to say distortion and attenuation introduced by the telephone connection, in accordance with the various frequencies. The data of that measurement are used by the central unit (14) to compensate for the signals which are successively received by way of the activated telephone connection. The system comprises a portable acoustic device (35) in which the digital data of the reference sound are recorded. That device may be connected to any telephone apparatus (12) of the network and comprises a manual control for triggering transmission to the central unit (14) of the sequence of reference sounds after the telephone connection has been established by means of the telephone apparatus (12 ). The portable device (35) may also have a recording of a code for identifying the user, for example, for use in a vocal postal system. In addition in a system in which the central unit (14) comprises a voice recognition device (17) the portable device (35) may have the recording of a series of phonetic prototypes, or phonemes, which are specific to the particular user, to be transferred to the central unit to improve the quality of recognition of the speech transmitted thereby. Alternatively the phonemes of different users may be recorded in a memory of the central unit (14) and be selected thereby on the basis of identification code.

Claims

1. A system for measuring and automatically compensating for the distortions of a connection between a telephone apparatus (12) and a central voice processing unit (14) for a telephone network comprising at least one telephone line (11), characterised in that the central unit (14) comprises measuring means (16) which are activatable after a telephone connection has been established on the said line (11), for measuring the distortion introduced in the transmission by the telephone line, and compensation means (30, 34) controlled by the measuring means (16) for compensating for the distortion.

2. A system according to claim 1, characterised in that the measuring means (16) comprise a memory (27) for recording a series of theoretical digital data relating to analysis of a sequence of reference sounds, a spectral sound analyser (21) for converting the sounds received into a series of effective digital data, and means (30) for comparing the theoretical digital data with the effective digital data obtained by the spectral analyser (31) upon the reception of a sound corresponding to the reference sound by way of the telephone line (11).

3. A system according to claim 2, characterised by means (35) for transmission of a sequence of reference sounds co-operating with the telephone apparatus (12) for transmitting the said sequence of sounds after the telephone apparatus (12) has established the telephone connection to the central unit (14).

4. A system according to claim 3, characterised in that the transmission means (35) comprises a memory (39) for recording the theoretical digital data, and a converter (41) for converting this data into the corresponding reference sounds.

5. A system according to claim 3 or 4, characterised in that the transmission means (35) comprise a portable acoustic device capable of being connected to any telephone apparatus (12) of the network, the portable device comprising a manual control member (46) for causing reading of the memory (39) and generation of the sequence of reference sounds which are then transmitted by means of the telephone apparatus (12) to which it is connected.

6. A system according to claim 5, characterised in that the portable device (35) comprises means (48) for signalling to the central unit (14) the identity of the portable device independently of the telephone apparatus to which it is connected.

7. A system according to claim 6, characterised in that the signalling means comprise an identification code recorded in the memory (48) and capable of being recognised by the central unit (14).

8. A system according to claim 6 characterised in that the portable device (35) is an acoustic or electromagnetic coupler which can be coupled to the microphone (43) of the telephone apparatus (12).

9. A system according to any of claim 6 wherein the central unit (14) is capable of controlling a voice post installation comprising for each user a post box (18) in which the voice messages addressed to the respective user can be recorded in digital form, characterised in that the central unit (14) is conditioned by the identification code to select the respective post box (18) in such a way as to make the voice message recorded therein available at the telephone apparatus (12) to which the portable device (35) is connected.

10. A system according to claim 2 characterised in that the spectral analyser (31) comprises an array of band pass filters (33), each being capable of isolating a corresponding frequency range, the measuring means (16) being capable of producing a curve in respect of the frequency response of the telephone line.

11. A system according to claim 10, characterised in that the sequence of reference sounds includes at least one period of silence, the measuring means (16) also being capable of measuring the strength of the noise generated by the line (11).

12. A system according to claim 11, characterised in that the compensating means (30, 34) are controlled by the measuring means (16) in such a way as to compensate both for the noise introduced by the telephone line (11) and the level and the frequency characteristics of the received sound.

13. A system according to claim 12, characterised in that the central unit (14) is capable of calculating both the energy of the noise received and the variation in energy of the acoustic level in each frequency of the sound received, the energy of the noise and the variation being stored (34) for the duration of the connection and being used to reconstruct the transfer function of the telephone line (11) during the connection.

14. A system according to claim 7, characterised in that the central unit (14) comprises a voice recognition device (17) having a memory (23) recorded with a series of phonetic prototypes to be compared to the signals of the received message, the compensating means (16) being capable of modifying the phonetic prototypes or the signals of the message received.

15. A system according to claim 14, characterised in that the phonetic prototypes are stored in the memory (49) of the portable device (35) and are transmitted to the recognition device (17) after the transmission of the identification code.

16. A system according to claim 14, characterised in that the memory (23) of the recognition device is recorded in separate regions with the phonetic prototypes of the various users, the central unit (14) being conditioned by the identification code to select on each occasion for the analysis of a message the memory region in which the phonetic prototypes corresponding to that identification code are recorded.

17. A system according to claim 14, characterised in that the memory (23) of the recognition device (17) is recorded with the phonetic prototypes of a plurality of voices, the central unit (14) being conditioned by the identification code to select on each occasion for analysis of a message the phonetic prototypes to be used for the user identified in that way.

18. A portable device (35) capable of being acoustically or electromagnetically coupled to a telephone apparatus (12) of a telephone network incorporating the compensation system according to claim 1 characterised in that the device (35) comprises a memory (39) recorded with the said digital data and a converter (41) for converting this data into the corresponding reference sounds.

19. A device according to claim 18, characterised in that it comprises a single control member (46) for causing reading of the memory (39).

20. A device according to claim 18 characterised in that the memory (39, 48) of the device (35) is also recorded with an identification code identifying the user carrying the device.

21. A device according to claim 18, characterised in that the memory (39, 49) of the device (35) is also recorded with a series of data corresponding to phonemes originally pronounced by the user of the device.