EP1439525A1 - Optimisation de distortion de transitions - Google Patents

Optimisation de distortion de transitions Download PDF

Info

Publication number
EP1439525A1
EP1439525A1 EP03000942A EP03000942A EP1439525A1 EP 1439525 A1 EP1439525 A1 EP 1439525A1 EP 03000942 A EP03000942 A EP 03000942A EP 03000942 A EP03000942 A EP 03000942A EP 1439525 A1 EP1439525 A1 EP 1439525A1
Authority
EP
European Patent Office
Prior art keywords
acoustic
parameters relating
sequences
calculated
communication network
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
EP03000942A
Other languages
German (de)
English (en)
Inventor
Bart Dr. Baekelandt
Lorenzo Guarino
Kurt Smet
Joeri Van Hoyweghen
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Siemens AG
Original Assignee
Siemens AG
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Siemens AG filed Critical Siemens AG
Priority to EP03000942A priority Critical patent/EP1439525A1/fr
Publication of EP1439525A1 publication Critical patent/EP1439525A1/fr
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
    • G10L19/04—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using predictive techniques
    • G10L19/16—Vocoder architecture
    • G10L19/18—Vocoders using multiple modes
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • G10L21/0316—Speech enhancement, e.g. noise reduction or echo cancellation by changing the amplitude
    • G10L21/0364—Speech enhancement, e.g. noise reduction or echo cancellation by changing the amplitude for improving intelligibility

Definitions

  • the invention relates to a method and an apparatus for Improve the quality of encoded acoustic sequences before storing the sequences in a storage unit in a communication network.
  • acoustic sequences in several transmission modes coded saved. These saved, encoded Sequences can be made from different tones or announcements with and to exist without breaks.
  • codec Modes coded saved.
  • These saved, encoded Sequences can be made from different tones or announcements with and to exist without breaks.
  • transition disorders Phase Phase, amplitude or frequency of the acoustic sequences.
  • the Transitional disorders that z. B. audible as "click" noises may affect the quality of the broadcast acoustic sequences.
  • the object of the present invention is a cost effective and easy to implement solution for Optimization of the data quality for coded, saved to get acoustic sequences.
  • a core of the invention is that Calculation of parameters relating to a transition disorder between coded sequences, the selection of the minimum Transition disturbance resulting parameters and the storage of the acoustic sequences in a table in one Communications network.
  • the solution is very cost effective and easy to implement.
  • Developments of the invention are specified in the subclaims. To the cache emptying a receiving decoder becomes the acoustic one Sequence a reset vector (e.g. homing frame) added. By emptying the decoder, the data quality of the acoustic sequence during transmission increased.
  • a reset vector e.g. homing frame
  • Figure 1 shows how for each transmission frame the cost function (cost function), which is used to calculate parameters relating to a transition disturbance between coded sequences, when transmitting acoustic sequences after encoding - decoding in e.g. B. a transcoding card as well as after reading stored coded sequences with subsequent decoding (lookup table for generating acoustic sequences in a communication network) is calculated.
  • the acoustic sequences can have different formats, such as. B.
  • CELP Code Excited Linear Prediction
  • ACELP Algebraic Code Excited Linear Prediction
  • MPEG-4 CELP MRWB-ACELP (Wideband Multi-Rate Algebraic Code Excited Linear Prediction)
  • G729, G723.1 or GSM-EFR Global Standard for Mobile Communications - Enhanced Full Rate.
  • the transmission energy can be dependent on the phase ( ⁇ ), the amplitude (a), the frequency (f) or the storage creation vector (v).
  • An nth transmission frame x n (a, f, ⁇ ) is encoded with a transmission mode in an encoder 1, the encoded transmission frame y n (a, f, ⁇ ) is then decoded with a decoder 2 and the transmission energy E n (a , f, ⁇ ) for calculating the cost function is passed on to a cost function unit 4.
  • a transmission frame y ' n (a, f, ⁇ ) is read out from a storage unit 3 and, in turn, the transmission energy E' n (a, f, ⁇ ) is passed on to the cost function unit 4 for calculating the cost function.
  • an energy difference function is calculated using the two transmission energies. Subsequently, the parameter relating to an optimization of the transition disturbance from the energy difference function searched with a selection unit 8.
  • the transition disturbance can be optimized according to a (amplitude) and / or f (frequency) and / or ⁇ (phase) and / or v (memory creation vector) happen.
  • the memory creation vector v describes the structure of the storage unit 3 for the Storage of acoustic sequences. So he parameterizes the data for the creation of such a storage unit 3.
  • the coded sequences with the parameter regarding one Optimization of the transition disorder are different Transmission modes (Codec Modes) are stored and therefore it must only the acoustic sequence when importing into a conversation selected, the compression of which is currently on the Voice channel used compression corresponds.
  • Figure 2 shows how the acoustic sequences from one Calculation and selection unit 9 for coding to one Encoder 1 can be forwarded.
  • the encoder receives the acoustic sequences via a receiving unit 5, conducts this for coding to a control unit 6 and sends it to a storage unit 3 by means of a transmission unit 7.
  • the acoustic sequences are in different Transmission modes and different period lengths saved.
  • the transmission modes are for transmission between a transmitter in the network and one Mobile station used over an air interface.
  • the too using transmission modes are previously between the Sending device in the network and the mobile station negotiated. Due to the transmission modes, the quality is the transmission of data packets improves and the error rate minimized.
  • AMR The language coding "AMR" was developed by the European Telecommunications Standardization Institute (ETSI) developed. AMR is standardized for GSM Cellular networks and is also used for the 3GPP standard. AMR was developed to ensure high voice quality in one broadband telecommunications network for the transmission of Language is guaranteed.
  • the AMR codec is a multi-mode Codec with 8 narrowband modes with bit rates between 4.75 and 12.2 kbps.
  • the sampling frequency is 8000 Hz and one Further processing takes place with 20 ms frames.
  • the AMR-WB Transmission mode is a multi-mode codec with 9 wideband Modes and bit rates between 6.6 and 23.85 kbps.
  • the Sampling frequency is 16000 Hz and further processing is also done with 20 ms frames.
  • a Storage unit 3 Before coding and later storage of the acoustic sequences in a Storage unit 3 becomes a parameter regarding a Optimization of the transition disorder to improve the quality of the acoustic sequences determined. Overage problems occur the interface of sequences if acoustic sequences repeated or linked to other acoustic sequences should be. The amplitude and / or frequency and / or phase must be adjusted so that there are no losses in quality become. Acoustic sequences contain at least one Frequency and can use frequency sequences with and / or as required to be without breaks. Each saved acoustic sequence will a reset vector (homing frame) attached. This sets the receiving decoder to a reset state, empties so its cache.
  • the reset vector is based on the last significant bits or a very small one Signal to be encoded.
  • the Saved encoded sequences are separated into individual frames divided and summarized to run (in) and Periodic (repetition) frames.
  • For storing acoustic sequences is only an administration of the ones to be transmitted acoustic sequences and no transcoding from the Original signal required, therefore, with minimal effort on a change of z.
  • the coded, acoustic Sequences are made using a transmitter sent to a mobile station.
  • FIG. 3 shows the result of the calculation of the cost function, in this case the energy difference function, for one Range of optimization size, such as here.
  • For the associated transition disorder becomes a minimal one Determines parameters and then the acoustic sequences encoded in an encoder 1 and in a storage unit 3 saved.
  • Figure 4 shows the calculation of the energy difference function per transmission frame of an acoustic sequence.
  • Figure 5 shows an example of a transition disorder.
  • the stringed acoustic sequences are not in one Phase and cause an audible "click" sound.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Quality & Reliability (AREA)
  • Transmission Systems Not Characterized By The Medium Used For Transmission (AREA)
EP03000942A 2003-01-16 2003-01-16 Optimisation de distortion de transitions Withdrawn EP1439525A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP03000942A EP1439525A1 (fr) 2003-01-16 2003-01-16 Optimisation de distortion de transitions

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
EP03000942A EP1439525A1 (fr) 2003-01-16 2003-01-16 Optimisation de distortion de transitions

Publications (1)

Publication Number Publication Date
EP1439525A1 true EP1439525A1 (fr) 2004-07-21

Family

ID=32524141

Family Applications (1)

Application Number Title Priority Date Filing Date
EP03000942A Withdrawn EP1439525A1 (fr) 2003-01-16 2003-01-16 Optimisation de distortion de transitions

Country Status (1)

Country Link
EP (1) EP1439525A1 (fr)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2266213A (en) * 1992-04-13 1993-10-20 Cambridge Algorithmica Ltd Digital signal coding
EP1005021A2 (fr) * 1998-11-25 2000-05-31 Matsushita Electric Industrial Co., Ltd. Procédé et dispositif d'extraction de paramètres source basés sur les formants, pour le codage et la synthèse de parole, utilisant une fonction de coût et un filtrage inverse
US6311154B1 (en) * 1998-12-30 2001-10-30 Nokia Mobile Phones Limited Adaptive windows for analysis-by-synthesis CELP-type speech coding
EP1189208A1 (fr) * 2000-09-19 2002-03-20 Nokia Corporation Détection d'erreurs de transmission dans un décodeur de parole

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2266213A (en) * 1992-04-13 1993-10-20 Cambridge Algorithmica Ltd Digital signal coding
EP1005021A2 (fr) * 1998-11-25 2000-05-31 Matsushita Electric Industrial Co., Ltd. Procédé et dispositif d'extraction de paramètres source basés sur les formants, pour le codage et la synthèse de parole, utilisant une fonction de coût et un filtrage inverse
US6311154B1 (en) * 1998-12-30 2001-10-30 Nokia Mobile Phones Limited Adaptive windows for analysis-by-synthesis CELP-type speech coding
EP1189208A1 (fr) * 2000-09-19 2002-03-20 Nokia Corporation Détection d'erreurs de transmission dans un décodeur de parole

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
VAINIO J ET AL: "GSM EFR based multi-rate codec family", ACOUSTICS, SPEECH AND SIGNAL PROCESSING, 1998. PROCEEDINGS OF THE 1998 IEEE INTERNATIONAL CONFERENCE ON SEATTLE, WA, USA 12-15 MAY 1998, NEW YORK, NY, USA,IEEE, US, 12 May 1998 (1998-05-12), pages 141 - 144, XP010279082, ISBN: 0-7803-4428-6 *

Similar Documents

Publication Publication Date Title
DE60121201T2 (de) Verfahren und vorrichtung zur verschleierung von fehlerhaften rahmen während der sprachdekodierung
DE60219351T2 (de) Signaländerungsverfahren zur effizienten kodierung von sprachsignalen
DE60031002T2 (de) Multimodaler mischbereich-sprachkodierer mit geschlossener regelschleife
DE69915830T2 (de) Verbesserte verfahren zur rückgewinnung verlorener datenrahmen für ein lpc-basiertes, parametrisches sprachkodierungsystem.
DE60120766T2 (de) Indizieren von impulspositionen und vorzeichen in algebraischen codebüchern zur codierung von breitbandsignalen
DE602004007786T2 (de) Verfahren und vorrichtung zur quantisierung des verstärkungsfaktors in einem breitbandsprachkodierer mit variabler bitrate
DE60132217T2 (de) Übertragungsfehler-verdeckung in einem audiosignal
DE60125219T2 (de) Spektralmerkmal ersatz für die verschleierung von rahmenfehlern in einem sprachdekoder
DE60123651T2 (de) Verfahren und vorrichtung zur robusten sprachklassifikation
DE60220485T2 (de) Verfahren und Vorrichtung zur Verschleierung von Rahmenausfall von prädiktionskodierter Sprache unter Verwendung von Extrapolation der Wellenform
DE602004003610T2 (de) Halbrätiger Vocoder
DE60316396T2 (de) Interoperable Sprachkodierung
DE60117144T2 (de) Sprachübertragungssystem und verfahren zur behandlung verlorener datenrahmen
DE60222445T2 (de) Verfahren zum verbergen von bitfehlern für die sprachcodierung
DE60024123T2 (de) Lpc-harmonischer sprachkodierer mit überrahmenformat
DE60129544T2 (de) Kompensationsverfahren bei rahmenauslöschung in einem sprachkodierer mit veränderlicher datenrate
DE60118631T2 (de) Verfahren zum ersetzen verfälschter audiodaten
CN102985969B (zh) 编码装置、解码装置和编码方法、解码方法
DE69923079T2 (de) Kodierung von stimmlosen sprachsegmenten mit niedriger datenrate
EP2062254B1 (fr) Stéganographie dans des codeurs de signaux numériques
DE69911169T2 (de) Verfahren zur dekodierung eines audiosignals mit korrektur von übertragungsfehlern
DE60224962T2 (de) Verfahren und Vorrichtung zur Verschleierung von fehlerbehafteten Sprachrahmen
DE60032006T2 (de) Prädiktionssprachkodierer mit musterauswahl für kodierungsshema zum reduzieren der empfindlichkeit für rahmenfehlern
DE602004012600T2 (de) Transcodierung zwischen den indizes von mehrimpuls-wörterbüchern zur codierung bei der digitalen signalkomprimierung
EP1080464B1 (fr) Procede et dispositif de codage de la parole

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LI LU MC NL PT SE SI SK TR

AX Request for extension of the european patent

Extension state: AL LT LV MK RO

AKX Designation fees paid
REG Reference to a national code

Ref country code: DE

Ref legal event code: 8566

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20050122