EP0094681A1 - Arrangement for electronic speech synthesis - Google Patents

Arrangement for electronic speech synthesis Download PDF

Info

Publication number
EP0094681A1
EP0094681A1 EP83104873A EP83104873A EP0094681A1 EP 0094681 A1 EP0094681 A1 EP 0094681A1 EP 83104873 A EP83104873 A EP 83104873A EP 83104873 A EP83104873 A EP 83104873A EP 0094681 A1 EP0094681 A1 EP 0094681A1
Authority
EP
European Patent Office
Prior art keywords
speech
filter
filters
individual
circuit arrangement
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.)
Granted
Application number
EP83104873A
Other languages
German (de)
French (fr)
Other versions
EP0094681B1 (en
Inventor
Hans Dipl.-Ing. Brandl
Werner Dipl.-Ing. Liegl
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 AT83104873T priority Critical patent/ATE26354T1/en
Publication of EP0094681A1 publication Critical patent/EP0094681A1/en
Application granted granted Critical
Publication of EP0094681B1 publication Critical patent/EP0094681B1/en
Expired 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
    • G10L13/00—Speech synthesis; Text to speech systems
    • G10L13/02—Methods for producing synthetic speech; Speech synthesisers
    • G10L13/04—Details of speech synthesis systems, e.g. synthesiser structure or memory management

Definitions

  • the invention relates to a circuit arrangement for electronic speech synthesis, speech elements being represented by significant parameters and individual speech elements being able to be put together to form longer speech segments, an excitation signal for representing voiced or unvoiced sounds by a tone or noise generator using at least some of the significant ones Parameters are generated and fed to a filter circuit and the filter output electrical signals are prepared for acoustic reproduction of the desired speech elements and segments.
  • the first group includes the methods in which the speech elements which were initially scanned in an analysis, possibly digitized and stored, for example, in a read-only memory, are retrieved from the memory and put together again for speech synthesis; redundant components that are not necessary for the intelligibility of the language are also stored, so that In this way, good quality speech per se can be generated, although there is a correspondingly high memory requirement for displaying an extensive vocabulary.
  • redundant speech components are largely hidden and the speech is stored in the form of significant parameters of speech elements, from which speech that is easy for the listener to understand can be generated with much less memory requirement.
  • the heart of known circuit arrangements for performing the latter method are filter circuits with variable filter coefficients.
  • a speech synthesis circuit is known in which an excitation signal to which significant speech parameters are applied is fed to a filter circuit with variable filter coefficients. These filter coefficients are further significant. Speech parameters are continuously controlled during the entire synthesis process, so that the circuit arrangement for carrying out these methods must have devices for storing these filter coefficients.
  • the circuit arrangement is to be provided with control devices for retrieving these coefficients from the memory and feeding them into the filters.
  • Such tunable filters have relatively large dimensions and are difficult and difficult to implement with the narrow tolerances required for good speech quality.
  • the invention is based on the object of providing a circuit arrangement for electronic speech synthesis which is easier to implement and which generates easily understandable speech.
  • the filter circuit consists of individual filters, of which only those individual filters, the filter coefficients of which represent the remaining significant parameters of the speech elements actually desired sequentially, are controlled sequentially.
  • the invention has the advantage of a lower memory requirement, combined with the advantage that filters with fixed coefficients can be implemented more easily.
  • individual filters constructed in analog technology can be provided which can be acted upon by a time-discrete analog excitation signal; the implementation of such analog filters is particularly simple. This applies all the more if they are expediently constructed in a further embodiment of the invention as a transversal filter using so-called CCD technology.
  • individual filters constructed in digital technology can also be provided, which can be loaded with a discrete-time digital excitation signal, which has the advantage of being able to store the parameter values of the speech signals in a particularly simple manner.
  • the individual filters can be addressed individually for the display of speech elements.
  • a circuit arrangement according to the invention can contain such a number of individual filters that all phonemes of a certain language can be represented. Several phonemes can be generated in a specific chronological order and connected to one another according to the characteristics of the human voice.
  • the individual filters can be interconnected in filter sets to represent longer language segments, with the filter sets being controlled by addressing specific to the filter set.
  • Such an alternative is characterized by lower memory requirements and is particularly suitable for the representation of speech in which the same speech segments occur repeatedly.
  • the individual filters can also be arranged in a matrix in which the individual filters of a matrix line are acted upon in parallel with the respective excitation signal, and the individual filter outputs of each matrix line can be connected sequentially to the matrix output.
  • the individual filters can be designed as linear prediction or formant filters.
  • The have fixed formant center frequencies and bandwidths.
  • Language elements are represented by rendering at least the three lowest formants.
  • the individual filters can be implemented using what is known as CCD technology. They can also be designed as a transversal filter.
  • FIG. 1 shows a block diagram of a circuit arrangement according to the invention, which essentially comprises a filter circuit F, an excitation generator device G and a control unit StE, which in turn are connected to an input unit EG or a low-pass filter TP and an electroacoustic transducer.
  • the input unit EG may output information about the speech elements to be synthesized to the control unit StE. This information can be entered, for example, using a keypad. Keypad information can also be entered. Likewise, information about speech elements to be synthesized can also be supplied from external systems in a system-conforming representation.
  • the control unit StE may include devices for the intermediate storage and processing of this information, for example in the so-called "handshake mode" with the filter circuit F, as well as memories in which speech parameters are stored.
  • two line connections can lead from the control unit StE to the filter circuit F, namely one for a filter change clock signal Tw and one for a digital speech element selection signal SEA, the filter change clock signal in the filter circuit F controls the synthesis of that speech element which is determined by the speech element selection signal SEA.
  • the e filter circuit F which can be designed, for example, in the manner shown in FIG. 2, contains, among other things, individual filters (F11 ... in FIG. 2) with fixed coefficients.
  • the filter circuit F can supply the control unit StE with a digital signal via a control line, which supply the control line E with a digital signal which indicates the synthesis process of a speech element in the handshake mode mentioned above, the information required for the synthesis process of the requests certain subsequent language elements through the information read in.
  • FIG 2 shows circuit-specific details of an embodiment of the filter circuit F, the excitation generator G and the StE.
  • the representation in FIG. 2 is based on an arrangement of the individual filters in a matrix.
  • the filter circuit F comprises, in addition to those in columns F11, F21,... Fn1; ... and lines F11, F12, ... F1Z; .... arranged individual filters, line-specific multiplexers M1, M2, ... Mn a line selection multiplexer ZMF, a selection circuit EME and optionally a swish excitation generator device not shown in FIG. 2 G and the time filter circuit inserted in the individual filters.
  • the excitation generator device G consists of a controllable pulse and noise generator IG or RG as well as a switching device.
  • the control unit StE includes, among other things, memories S1, Sn, in which speech parameter values are stored, a filter alternating clock generator FwG, and one. Memory selection circuit ZMA.
  • the filter circuit F receives a filter change clock pulse Tw and a speech element selection signal SEA from the control unit StE.
  • the filter change clock generator FwG arranged in the control unit Ste generates equidistant filter change act pulses with a pulse period Tw, which can be, for example, between 10 and 25 ms.
  • the filter change clock pulses are simultaneously fed to all line-specific multiplexers M1 ...
  • the number of line-specific multiplexers M is equal to the number of said memories S. This number corresponds to the number of lines in the filter matrix. If, as assumed above, different language segments can be generated by the filter circuit F n , the filter circuit F n has filter sets arranged in rows. Each filter set comprises at least one individual filter. The language segment generated in the filter set can be composed of several language elements which are generated in the filters belonging to this filter set. The duration of a language element is t W ; the duration of a language segment composed of m language elements is then mt w .
  • the number of fin filters required for the generation of such a speech segment can be less than m if the speech segment in question contains the same speech elements that are synthesized in the same individual filters of the relevant filter set.
  • the (analog) speech element signals are through interconnected the respective line-specific multiplexer M in the filter change cycle Tw to form the (analog) speech signal SA.
  • the pulse sequence with the frequency 1 / t w generated by the filter alternating clock generator FWG is also fed to all line-specific memories S in the control unit StE, in which the parameter values of the excitation signals, for example their frequency f and amplitude U, are stored. These parameter values are retrieved from the memories S and supplied to the memory selection circuit ZMA by the pulse sequence generated before the filter change clock generator FwG.
  • the latter selects the parameter values of the speech segment to be generated in accordance with the speech element selection signal SEA, which is also supplied to it, and feeds them to the controllable excitation generator device G.
  • the excitation generator device G comprises a pulse generator IG which can be controlled in frequency and amplitude, and a noise generator RG which can be controlled in amplitude.
  • the switching device provided on the output side in the excitation generator device G is controlled by the memory S via information about the frequency: for frequency values equal to zero, the noise generator RG is coupled to the filter circuit F, for frequency values not equal to zero the pulse generator IG is coupled to the filter circuit F.
  • the excitation generator device G delivers pulse or noise signals of a certain amplitude and, if appropriate, frequency. Unvoiced speech elements are simulated by noise signals, voiced speech elements of a certain frequency by pulse trains of precisely this frequency.
  • the excitation signals generated by the controllable excitation generator device G are supplied to all filters, including those which are not used to generate the selected speech segment. All created in the filter sets (Analog) signals are fed via the line-specific multiplexers M to the line selection multiplexer ZMF, in which the desired speech signal is selected by means of the speech element selection signal SGA, which then appears at the output SA of the line selection multiplexer.
  • the voice signal SA is fed to the low-pass filter TP, which filters out higher frequency components contained in the voice signal, for example due to pulse-like excitation of the filters.
  • the above explanations are not limited to a speech signal synthesis by means of analog filters loaded by analog excitation signals, but also apply in a corresponding manner to a speech signal application means by means of digital filters loaded by digital excitation signals, in which case the filter output signal is then a digital-analog signal. Undergoes change. After any amplification that may still be required, the electrical analog voice signal is finally reproduced via an electro-acistic converter.
  • the line-specific multiplexers M simultaneously emit a digital signal E to the selection circuit ZME when the last speech element generated in the filter set concerned is switched on, which digital signal identifies the completion of the speech synthesis process in the filter set.
  • the selection circuit ZME brought from the current speech element selection signal SEA into a corresponding switching position now switches the corresponding digital signal through to the control unit StE, which can thus trigger the synthesis process of the next speech segment.
  • the individual filters can also be individually addressable with fixed coefficients.
  • the filter circuit (F in FIG 2) no line selection multiplexer (ZMF in FIG. 2) and no selection circuit (ZME in FIG. 2) and no line-individual multiplexers (M in FIG. 2) of the type described above.
  • the control unit (StE in FIG. 2) then has devices for storing individually addressable parameter values for the excitation signals and for interconnecting those in the individual filters. producible speech element signals.
  • the filter change clock generator FwG (in FIG. 2) and the controllable excitation generator device G (in FIG. 2) (possibly with a time window circuit) also perform the functions described above in this embodiment.
  • Such an embodiment which is characterized by optional control of individual filters by means of individual filter-specific addressing vidual addressing, need only have different individual filters .., while in the exemplary embodiment according to FIG. 2 the same individual filters are also used in the different filter sets, possibly also in one and the same filter set can be arranged.
  • the latter alternatives which due to the addressing of the individual filter set compared to the addressing of the individual filter can be implemented with less control technology, are particularly suitable for the re-outlay, and are particularly suitable for the reproduction of language which repeatedly contains the same language segments.
  • Embodiments that contain both individual filters connected in filter sets and independent individual filters will be of practical importance. In this way, the number of individual filters used and the necessary tax expenditure can be optimized.
  • the individual filters used according to the invention can also be designed as linear prediction filters with fixed n coefficients according to FIG. 3.
  • Linear prediction tone as such is known and in the relevant specialist literature (for example FLANAGAN, Speech Analysis Synthesis and Perception, Springer-Verlag Berlin, Heidelberg,
  • the speech quality that can be achieved is proportional to the number of coefficients within certain limits. Good speech quality can be achieved with around 10 filter coefficients.
  • Such a prediction filter is connected with its connection terminals designated in FIG. 3 with A11 and B11 to the correspondingly designated connection terminals of the filter circuit F in accordance with FIG. 2 in order to form the individual filter F11 there.
  • the linear prediction filters used according to the invention can be designed as analog or digital filters. Accordingly, the filters are supplied with excitation signals in analog or digital form from the excitation generator; accordingly, analog or digital signals result at the filter outputs.
  • the individual filters used according to the invention can, as shown in FIG. 4, also be designed as formant filters with fixed filter coefficients, wherein each individual filter can correspond to a parallel connection of three formant filters simulating at least the first three (low-frequency) speech novels.
  • the generation of speech by formant synthesis is known and adequately described (for example in the literature FLANAGAN p. 339 ff. Already cited), so that there is no need to describe it here.
  • the formant filters are expediently designed as bandpasses with certain fixed pass bands and center frequencies of these pass bands.
  • Such a filter circuit can also be implemented in analog and digital technology.
  • the individual filters can be implemented in so-called CCD technology in all of the above cases.
  • the excitation signal is fed to the individual filters in a time-discrete form.
  • the filter circuit (F in FIG. 2) can contain a time window circuit not shown in FIG. 2.
  • the time window circuit can generate a sampling signal of fixed frequency, which has at least twice the frequency with respect to the network signal to be sampled.
  • the controllable excitation generator device G and all individual filters are clocked into the filter circuit F with this scanning signal generated in this way.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Electrophonic Musical Instruments (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)
  • Photoreceptors In Electrophotography (AREA)
  • Machine Translation (AREA)
  • Toys (AREA)

Abstract

A circuit for electronically synthesizing speech has an audio generator for representing voiced sounds and a noise generator for representing voiceless sounds and a means for selecting significant parameters of the various speech elements by sampling and a means for storing those parameters. The circuit also includes a filter unit comprised of a number of individual filters and a means for selectively driving only those individual filters having filter coefficients necessary for representing the significant parameters of the particular speech element to be synthesized. The filters can be utilized individually or combined into selected groups in order to generate longer speech segments. The electronic signal at the output of the filter unit is edited for acoustically reproducing the desired speech elements and segments.

Description

Die Erfindung betrifft eine Schaltungsanordnung zur elektronischen Sprachsynthese, wobei Sprachelemente durch signifikante Parameter dargestellt werden und einzelne Sprachelemente zu längeren Sprachsegmenten zusammengesetzt werden können, wobei ein Anregungssignal zur Darstellung stimmhafter bzw. stimmloser Laute durch einen Ton- bzw. Rauschgenerator unter Verwendung mindestens eines Teils der signifikanten Parameter erzeugt und einer Filterschaltung zugeführt wird und wobei die filterausgangsseitigen elektrischen Signale zur akustischen Wiedergabe der gewünschten Sprachelemente und -segmente aufbereitet werden.The invention relates to a circuit arrangement for electronic speech synthesis, speech elements being represented by significant parameters and individual speech elements being able to be put together to form longer speech segments, an excitation signal for representing voiced or unvoiced sounds by a tone or noise generator using at least some of the significant ones Parameters are generated and fed to a filter circuit and the filter output electrical signals are prepared for acoustic reproduction of the desired speech elements and segments.

Es sind bereits Verfahren zur Erzeugung von Sprachelementen bekannt, die zur Bildung von längeren Sprachsegmenten zusammengesetzt werden können; diese Verfahren lassen sich in die folgenden zwei Gruppen einteilen. Zur ersten Gruppe gehören die Verfahren, bei denen die zunächst in einer Analyse abgetasteten, eventuell digitalisierten und beispielsweise in einem Festwertspeicher abgespeicherten Sprachelemente zur Sprachsynthese aus dem Speicher wieder abgerufen und zusammengesetzt werden; dabei werden auch redundante, für die Verständlichkeit der Sprache nicht notwendige Bestandteile abgespeichert, so daB auf diese Weise an sich Sprache guter Qualität erzeugt werden kann, wobei zur Darstellung eines umfangreichen Wortschatzes allerdings ein entsprechend hoher Speicherbedarf besteht. Bei der zweiten Gruppe der Sprachsyntheseverfahren sind redundante Sprachbestandteile weitgehend ausgeblendet und die Sprache ist in Form von signifikanten Parametern von Sprachelementen gespeichert, aus denen sich so bei wesentlich geringerem Speicherbedarf für den Hörer gut verständliche Sprache erzeugen läßt. Das Kernstück bekannter Schaltungsanordnungen für die Durchführung der letztgenannten Verfahren bilden Filterschaltungen mit variablen Filter-Koeffizienten. So ist beispielsweise aus der DE-AS 22 09 548 eine Sprachsyntheseschaltung bekannt, bei der ein mit signifikanten Sprachparametern beaufschlagtes Anregungssignal einer Filterschaltung mit variablen Filterkoeffizienten zugeführt wird. Diese Filterkoeffizienten werden durch weitere signifikante. Sprachparameter fortlauf erd während des gesamten Synthesevorganges gesteuert, so daß die Schaltungsanordnung zur Durchführung dieser Verfahren Einrichtungen zur Speicherung eben dieser Filterkoeffizienten aufzuweisen hat. Außerdem ist die Schaltungsanordnung mit Steuereinrichtungen zum Abrufen dieser Koeffizienten aus dem Speicher und ihrer Zuführung in die Filter zu versehen. Solche abstimmbare Filter weisen dabei relativ große Abmessungen auf und lassen sich mit den für gute Sprachqualität notwendigen engen Toleranzen nur schwer und nur unter großem Aufwand realisieren.Methods for generating language elements are already known which can be put together to form longer language segments; these methods can be divided into the following two groups. The first group includes the methods in which the speech elements which were initially scanned in an analysis, possibly digitized and stored, for example, in a read-only memory, are retrieved from the memory and put together again for speech synthesis; redundant components that are not necessary for the intelligibility of the language are also stored, so that In this way, good quality speech per se can be generated, although there is a correspondingly high memory requirement for displaying an extensive vocabulary. In the second group of speech synthesis methods, redundant speech components are largely hidden and the speech is stored in the form of significant parameters of speech elements, from which speech that is easy for the listener to understand can be generated with much less memory requirement. The heart of known circuit arrangements for performing the latter method are filter circuits with variable filter coefficients. For example, from DE-AS 22 09 548 a speech synthesis circuit is known in which an excitation signal to which significant speech parameters are applied is fed to a filter circuit with variable filter coefficients. These filter coefficients are further significant. Speech parameters are continuously controlled during the entire synthesis process, so that the circuit arrangement for carrying out these methods must have devices for storing these filter coefficients. In addition, the circuit arrangement is to be provided with control devices for retrieving these coefficients from the memory and feeding them into the filters. Such tunable filters have relatively large dimensions and are difficult and difficult to implement with the narrow tolerances required for good speech quality.

Der Erfindung liegt nun die Aufgabe zugrunde, eine einfacher realisierbare Schaltungsanordnung zur elektronischen Sprachsynthese zu schaffen, die gut verständliche Sprache erzeugt.The invention is based on the object of providing a circuit arrangement for electronic speech synthesis which is easier to implement and which generates easily understandable speech.

Die Aufgabe wird erfindungsgemäß dadurch gelöst, daß die Filterschaltung aus Einzelfiltern besteht, von denen jeweils nur diejenigen Einzelfilter, deren Filterkoeffizienten die restlichen signifikanten Parameter der tatsächlich sequentiell gewünschten Sprachelemente darstellen, sequentiell angesteuert werden. Die Erfindung bringt den Vorteil eines geringeren Speicheraufwandes mit sich, verbunden mit dem Vorteil einer einfacheren Realisierbarkeit von Filtern mit festen Koeffizienten.The object is achieved in that the filter circuit consists of individual filters, of which only those individual filters, the filter coefficients of which represent the remaining significant parameters of the speech elements actually desired sequentially, are controlled sequentially. The invention has the advantage of a lower memory requirement, combined with the advantage that filters with fixed coefficients can be implemented more easily.

Weitere vorteilhafte Ausgestaltungen der Erfindungen ergeben sich aus den Unteransprüchen. So können in Analogtechnik aufgebaute Einzelfilter vorgesehen sein, die mit einem zeitdiskreten Analog-Anregungssignal beaufschlagbar sind; die Realisierung solcher Analogfilter ist besonders einfach. Dies gilt um so mehr, wenn sie zweckmäßigerweise in einer weiteren Ausgestaltung der Erfindung als Transversalfilter in sogenannter CCD-Technik aufgebaut sind. Alternativ können auch in Digitaltechnik aufgebaute Einzelfilter vorgesehen sein,
die mit einem zeitdiskreten Digital-Anregungssignal beaufschlagbar sind, was den Vorteil mit sich bringt, die Parameterwerte der Sprachsignale in besonders einfacher Weise abspeichern zu können.Further advantageous refinements of the inventions result from the subclaims. Thus, individual filters constructed in analog technology can be provided which can be acted upon by a time-discrete analog excitation signal; the implementation of such analog filters is particularly simple. This applies all the more if they are expediently constructed in a further embodiment of the invention as a transversal filter using so-called CCD technology. Alternatively, individual filters constructed in digital technology can also be provided,
which can be loaded with a discrete-time digital excitation signal, which has the advantage of being able to store the parameter values of the speech signals in a particularly simple manner.

In weiterer Ausgestaltung der Erfindung können die Einzelfilter zur Darstellung von Sprachelementen filterindividuell adressiert werden. Eine Schaltungsanordnung gemäß der Erfindung kann eine solche Anzahl von Einzelfiltern enthalten, daß alle Phoneme einer bestimmten Sprache darstellbar sind. Mehrere Phoneme können in bestimmter zeitlicher Reihenfolge erzeugt und entsprechend der Charakteristik der menschlichen Stimme miteinander verbunden werden.In a further embodiment of the invention, the individual filters can be addressed individually for the display of speech elements. A circuit arrangement according to the invention can contain such a number of individual filters that all phonemes of a certain language can be represented. Several phonemes can be generated in a specific chronological order and connected to one another according to the characteristics of the human voice.

In weiterer Ausgestaltung der Erfindung können die Einzelfilter in Filtersätzen zur Darstellung längerer Sprachsegmente zusammengeschaltet sein, wobei eine wahlfreie Ansteuerung der Filtersätze durch filtersatzindividuelle Adressierung erfolgt. Eine solche Alternative zeichnet sich durch geringeren Speicheraufwand aus und ist besonders für die Darstellung von Sprache geeignet, in der gleiche Sprachsegmente wiederholt auftreten. Die Einzelfilter können auch in einer Matrix angeordnet sein, in der jeweils die Einzelfilter einer Matrixzeile parallel mit dem jeweiligen Anregungssignal beaufschlagt werden, und die Einzelfilterausgänge jeweils einer Matrixzeile sequentiell mit dem Matrixausgang verbindbar sind.In a further embodiment of the invention, the individual filters can be interconnected in filter sets to represent longer language segments, with the filter sets being controlled by addressing specific to the filter set. Such an alternative is characterized by lower memory requirements and is particularly suitable for the representation of speech in which the same speech segments occur repeatedly. The individual filters can also be arranged in a matrix in which the individual filters of a matrix line are acted upon in parallel with the respective excitation signal, and the individual filter outputs of each matrix line can be connected sequentially to the matrix output.

In weiterer Ausgestaltung der Erfindung können die Einzelfilter als Linearprädiktions- bzw. als Formantfilter ausgebildet sein. Die weisen feste Formantmittenfrequenzen und -bandbreiten auf. Die Darstellung von Sprachelementen erfolgt dabei durch die Wiedergabe mindestens der drei tiefsten Formanten.In a further embodiment of the invention, the individual filters can be designed as linear prediction or formant filters. The have fixed formant center frequencies and bandwidths. Language elements are represented by rendering at least the three lowest formants.

Die Einzelfilter können in einer weiteren Ausgestaltung der Erfindung in sogenannter CCD-Technik realisiert sein. Auch können sie als Transversalfilter ausgebildet sein.In a further embodiment of the invention, the individual filters can be implemented using what is known as CCD technology. They can also be designed as a transversal filter.

Die Erfindung wird nun in einem zum Verständnis erforderlichen Umfang anhand von Zeichnungen näher beschrieben. Dabei zeigt

  • FIG 1 ein Blockschaltbild einer Schaltungsanordnung zur Sprachsynthese.
  • FIG 2 zeigt das Blockschaltbild eines Ausführungsbeispiels der Erfindung mit in Matrixform angeordneten Einzelfiltern.
  • FIG 3 und FIG 4 zeigen die Verwendung von Linearprädiktions- bzw. Formantfiltern.
The invention will now be described in more detail to the extent necessary for understanding with reference to drawings. It shows
  • 1 shows a block diagram of a circuit arrangement for speech synthesis.
  • 2 shows the block diagram of an embodiment of the invention with individual filters arranged in matrix form.
  • 3 and 4 show the use of linear prediction or formant filters.

FIG 1 zeigt ein Blockschaltbild einer Schaltungsanordnung gemäβ der Erfindung, die im Kern eine Filterschaltung F, eine Anregungsgeneratoreinrichtung G und eine Steuereinheit StE umfaβt, die ihrerseits mit einer Eingabeeinheit EG bzw. mit einem Tiefpaβ TP und einem elektroakustishen Wandler in Verbindung stenen. Die Eingabeeinheit EG möge Informationen über zu synthetisierende Sprachelemente an die Steuereinheit StE ab- geben. Diese Informationen können beispielsweise über ein Tastenfeld eingegeben werden. Ebenso können Informationen Tastenfeld eingegeben werden. Ebenso können Informationen über zu synthetisierende Sprachelemente aber auch aus systemexternen Einrichtungen in systemkonformer Dar- stellung zugeführt werden. Die Steuereinheit StE möge Einrichtungen zur Zwischenspeicherung und Abarbeitung dieser Informationen z.B. im sogenannten "handshake-Betrieb" mit der Filterschaltung F sowie Speicher umfassen, in denen Sprachparameter abgespeichert sind. Wie in FIG 1 angedeutet ist, können von der Steuereinheit StE zu der Filterschaltung F beispielsweise zwei Leitungsverbindungen führen, nämlich eine für ein Filterwechseltaktsignal Tw und eine für ein digitales Sprachelement- auswahlsignal SEA, wobei das Filterwechseltaktsignal in der Filterschaltung F dieSynthese desjenigen Sprach- elements steuert, das durch das Sprachelementauswahlsignal SEA bestimmt wird. Die e Filterschaltung F, die beispielsweise in der aus FIG 2 dargestellten Weise ausge- bildet sein kann, enthält unter anderem Einzelfilter (F11... in FIG 2)mit festen Koeffizienten. Mit Hilfe dieser Einzelfilter wird die Sprachsynthese durchgeführt, wobei die Einzelfilter einem Ausgang SA der Filtersphaltung F ein (elektrishes) Sprachsignal liefern, das gegebenenfalls nach Digital/Analog-Wandlung über ein das gegebenenfalls nach Digital/Analog-Wandlung über ein Tiefpaβfilter TP und gegebenenfalls über einen nach- folgendenVerstärker einem elektroakustischen Wandler zugefükut wird. Wie in FIG 1 weiter angedeutet ist, kann die Filterschaltung F der Steuereinheit StE über eine Steuerleitung ein Digitalsignal zuführen, das das Steuerleitung E ein Digitalsignal zuführen, das das Synthesevorgangs eines Sprachelementsanzeigt im sereits erwähnten handshake-Betrieb die erforder- lichen Informationen für den Synthesevorgang des durch die eingegesene Information bestimmten nachfolgenden Sprach-elements anfordert.1 shows a block diagram of a circuit arrangement according to the invention, which essentially comprises a filter circuit F, an excitation generator device G and a control unit StE, which in turn are connected to an input unit EG or a low-pass filter TP and an electroacoustic transducer. The input unit EG may output information about the speech elements to be synthesized to the control unit StE. This information can be entered, for example, using a keypad. Keypad information can also be entered. Likewise, information about speech elements to be synthesized can also be supplied from external systems in a system-conforming representation. The control unit StE may include devices for the intermediate storage and processing of this information, for example in the so-called "handshake mode" with the filter circuit F, as well as memories in which speech parameters are stored. As indicated in FIG. 1, two line connections can lead from the control unit StE to the filter circuit F, namely one for a filter change clock signal Tw and one for a digital speech element selection signal SEA, the filter change clock signal in the filter circuit F controls the synthesis of that speech element which is determined by the speech element selection signal SEA. The e filter circuit F, which can be designed, for example, in the manner shown in FIG. 2, contains, among other things, individual filters (F11 ... in FIG. 2) with fixed coefficients. With the aid of these individual filters, the speech synthesis is carried out, the individual filters supplying an output SA of the filter circuit F an (electrical) speech signal which, if appropriate after digital / analog conversion via a possibly after digital / analog conversion via a low-pass filter TP and possibly via a subsequent amplifier is fed to an electroacoustic transducer. As is further indicated in FIG. 1, the filter circuit F can supply the control unit StE with a digital signal via a control line, which supply the control line E with a digital signal which indicates the synthesis process of a speech element in the handshake mode mentioned above, the information required for the synthesis process of the requests certain subsequent language elements through the information read in.

FIG 2 seigt schaltungstechnische Einzelheiten eines Ausführungsbeispiels der Filterchaltung F, der Anregungsgenera einrichtung G sowie der StE.2 shows circuit-specific details of an embodiment of the filter circuit F, the excitation generator G and the StE.

Dabei geht die Darstellung in FIG 2 von einer Anordnung der Einzelfilter in matrix aus.Die Filterschaltung F umfaβt dabei neben den in Spalten F11, F21,...Fn1; ... und Zeilen F11, F12, ...F1Z;.... angeordneten Einzelfiltern, zeilenindividuelle Multiplexer M1, M2,...Mn einen Zeilenauswahlmultiplexer ZMF, eine Auswahlchaltung EME sowie gegebenenfalls eine in FIG 2 nicht näher dargestellte, swischen Anregungsgeneratoreinrichtung G und den Einselfiltern eingefügten Zeitfensterschaltung.The representation in FIG. 2 is based on an arrangement of the individual filters in a matrix. The filter circuit F comprises, in addition to those in columns F11, F21,... Fn1; ... and lines F11, F12, ... F1Z; .... arranged individual filters, line-specific multiplexers M1, M2, ... Mn a line selection multiplexer ZMF, a selection circuit EME and optionally a swish excitation generator device not shown in FIG. 2 G and the time filter circuit inserted in the individual filters.

Die Anregungsgeneratoreinrichtung G besteht gemäβ FIG 2 aus je einem steuerbaren Impuls- und Rauschgenerator IG bzw. RG sowie einer Schaltvorrichtung Die Steuereinheit StE umfaßt unter anderem Speicher S1, Sn, in denen Sprachparameterwerte abgespeichert sind, einen Filterwechseltaktgenerator FwG, sowie eine. Speicherauswahlschaltung ZMA. Die Filterschaltung F erhält von der Steuereinheit StE einen Filterwechsel-Taktpuls Tw und ein Sprachelementauswahlsignal SEA. Der in der Steuereinheit Ste angeordnete Filterwechseltaktgenerator FwG erzeugt äquidistante Filterwechseleaktimpulse mit einer Impulsperiode Tw, die beispielsweise zwischen 10 und 25 ms betragen kann. Die Filterwechseltaktimpulse werden allen in der Filterschaltung F befindlichen zeilenindividuellen Multiplexern M1... und den in der Steuereinheit StE angeordneten Speichern S1...gleichzeitig zugeführt. Bei dem hier betrachteten Ausführungsbeispiel ist die Anzahl der zeilenindividuellen Multiplexer M gleich der Anzahl der genannten Speicher S. Diese Zahl entspricht der Zeilenzahl der Filtermatrix. Wenn - wie oben unterstelltdurch die Filterschaltung F n unterschiedliche Sprachsegmente erzeugbar sind, so weist die Filterschaltung Fn zeilenweise angeordnete Filtersätze auf. Jeder Filtersatz umfaßt mindestens ein Einzelfilter Das im Filtersatz erzeugte Sprachsegment kann aus mehreren Sprachelementen zusammengesetzt werden, die in den zu diesem Filtersatz gehörenden Filtern erzeugt werden. Die Dauer eines Sprachelementes beträgt tW; die Dauer eines aus m Sprachelementen zusammengesetzten Sprachsegments beträgt dann m.tw. Die Anzahl der für die Erzeugung eines solchen Sprachsegments erforderlichen Finzelfilter kann kleiner als m sein, wenn das betreffende Sprachsegment unterein- ander gleiche Sprachelemente enthält, die in gleichen Einzelfiltern des betreffenden Filtersatzes synthetisiert werden. Die (analogen) Sprachelementsignale werden durch den jeweiligen zeilenindividuellen Multiplexer M im Filterwechseltakt Tw zum (analogen) Sprachsignal SA zusammengeschaltet. Die vom Filterwechseltaktgenerator FWG erzeugte Impulsfolge mit der Frequenz 1/tw wird auch allen in der Steuereinheit StE befindlichen, zeilenindividuellen Speichern S zugeführt, in denen die Parameterwerte der Anregungssignale, Z.B. deren Frequenz f und Amplitude U abgespeichert sind. Durch die vor Filterwechseltaktgenerator FwG erzeugte Impulsfolge werden diese Parameterwerte aus den Speichern S abgerufen und der Speicherauswahlschaltung ZMA zugeführt. Diese wählt dann nach Maß- gabe des ihr ebenfalls zugeführten Sprachelementauswahlsignals SEA die Parameterwerte des zu erzeugenden Sprachsegments aus und führt diese der steuerbaren Anregungsgeneratoreinrichtung G zu. Die Anregungsgeneratoreinrichtung G umfaßt einen in Frequenz und Amplitude steuerbaren Impulsgenerator IG sowie einen in der Amplitude steuer― baren Rauschgenerator RG. Die in der Anregungsgenerator- einrichtung G ausgangasseitig vorgesehene Schaltvorrichtung wird durch den Speichern S abgerufene Information über die Frecuenz gesteuert: Für Frecuenzwerte gleich Null wird der Rauschgenerator RG an die Filterschaltung F gekoppelt, für Frequenzwert ungleich Null wird der Impulsgenerator IG an die Filterschaltung F gekoppelt. Die Anregungsgeneratcreinrichtung G liefert abhängig von den Parameterwerten f und U Impuls- oder Rauschsignale bestimmter Amplitude und ggf. Frequenz. Stimmlose Sprachelemente werden durch Rauschsignale, stimmhafte Sprachelemente bestimmter Frequenz durch Impulsfolgen eben dieser Frequenz nachgebildet. Die von der steuerbaren Anregungsgeneratoreinrichtung G erzeugten Anregungs- signale werden bei dem hier beschriebenen Ausführungs- beispiel allen Filter sälzen zugeführt, also auch denjenigen, die nicht der Erzeugung des ausgewählten Sprach- segments diener. Alle in den Filtersätzen erzeugten (Analog) signale werden über die zeilenindividuellen Multiplexer M dem Zeilenauswahlsultiplexer ZMF zugeführt, in dem mittels des Sprachelementauswahlsignals SGA das gewünschte Sprachsignal ausgewählt wird, das dann am Ausgang SA des Zeilenauswahlmultiplexers auftritt. Das Sprachsignal SA wird dem Tiefpaß TP zugeführt, der etwa auf Grund ispulsartiger Anregung der Filter im Sprachsignal enthaltene höhere Frequenzbestandteile ausfiltert. Es sei hier bemerkt, daß die vorstehenden Erläuterungen nicht auf eine Sprachsignalsynthese mittels durch Analog- Anregungssignale beaufschlagte Analogfilter beschränkt sind, sondern in entsprechender Weise auch für eine Sprachsignalanadrungmittels durch Digital-Anregungssignale beaufschlagte Digitalfilter gelten, wobei dann das Filterausgangssignal noch einer,Digital-Analog-Unwandlung unterzogen wird. Nach einer gegebenenfalls noch erforderlichen Verstärkung wird schließlich das elektrische Analog-Sprachsignal, über einen elektroakistischen Wandler wiedergegeben.According to FIG. 2, the excitation generator device G consists of a controllable pulse and noise generator IG or RG as well as a switching device. The control unit StE includes, among other things, memories S1, Sn, in which speech parameter values are stored, a filter alternating clock generator FwG, and one. Memory selection circuit ZMA. The filter circuit F receives a filter change clock pulse Tw and a speech element selection signal SEA from the control unit StE. The filter change clock generator FwG arranged in the control unit Ste generates equidistant filter change act pulses with a pulse period Tw, which can be, for example, between 10 and 25 ms. The filter change clock pulses are simultaneously fed to all line-specific multiplexers M1 ... in the filter circuit F and the memories S1 ... arranged in the control unit StE. In the exemplary embodiment considered here, the number of line-specific multiplexers M is equal to the number of said memories S. This number corresponds to the number of lines in the filter matrix. If, as assumed above, different language segments can be generated by the filter circuit F n , the filter circuit F n has filter sets arranged in rows. Each filter set comprises at least one individual filter. The language segment generated in the filter set can be composed of several language elements which are generated in the filters belonging to this filter set. The duration of a language element is t W ; the duration of a language segment composed of m language elements is then mt w . The number of fin filters required for the generation of such a speech segment can be less than m if the speech segment in question contains the same speech elements that are synthesized in the same individual filters of the relevant filter set. The (analog) speech element signals are through interconnected the respective line-specific multiplexer M in the filter change cycle Tw to form the (analog) speech signal SA. The pulse sequence with the frequency 1 / t w generated by the filter alternating clock generator FWG is also fed to all line-specific memories S in the control unit StE, in which the parameter values of the excitation signals, for example their frequency f and amplitude U, are stored. These parameter values are retrieved from the memories S and supplied to the memory selection circuit ZMA by the pulse sequence generated before the filter change clock generator FwG. The latter then selects the parameter values of the speech segment to be generated in accordance with the speech element selection signal SEA, which is also supplied to it, and feeds them to the controllable excitation generator device G. The excitation generator device G comprises a pulse generator IG which can be controlled in frequency and amplitude, and a noise generator RG which can be controlled in amplitude. The switching device provided on the output side in the excitation generator device G is controlled by the memory S via information about the frequency: for frequency values equal to zero, the noise generator RG is coupled to the filter circuit F, for frequency values not equal to zero the pulse generator IG is coupled to the filter circuit F. Depending on the parameter values f and U, the excitation generator device G delivers pulse or noise signals of a certain amplitude and, if appropriate, frequency. Unvoiced speech elements are simulated by noise signals, voiced speech elements of a certain frequency by pulse trains of precisely this frequency. In the exemplary embodiment described here, the excitation signals generated by the controllable excitation generator device G are supplied to all filters, including those which are not used to generate the selected speech segment. All created in the filter sets (Analog) signals are fed via the line-specific multiplexers M to the line selection multiplexer ZMF, in which the desired speech signal is selected by means of the speech element selection signal SGA, which then appears at the output SA of the line selection multiplexer. The voice signal SA is fed to the low-pass filter TP, which filters out higher frequency components contained in the voice signal, for example due to pulse-like excitation of the filters. It should be noted here that the above explanations are not limited to a speech signal synthesis by means of analog filters loaded by analog excitation signals, but also apply in a corresponding manner to a speech signal application means by means of digital filters loaded by digital excitation signals, in which case the filter output signal is then a digital-analog signal. Undergoes change. After any amplification that may still be required, the electrical analog voice signal is finally reproduced via an electro-acistic converter.

Die zeilenindividuellen Multiplexer M begen gleichzeitig mit dem Durchschalten des letzten in dem betreffenden Filtersatz erzeugten Sprachelements ein Digitalsignal E an die Auswahlschaltung ZME ab, das den zeitlichen Abschluß des Sprachsynthesevorgangs in dem Filtersatz kennzeichnet. Die ebenso wie der Zeilenauswahlmultiplexer ZMF von dem momentanen Sprachelementauswahlsignal SEA in eine entsprechende Schaltstllung gebrachte Auswahlschaltung ZME schaltet nun das entsprechende Digitalsignal zu der Steuereinheit StE durch, die damit den Synthesevorgang des nächstfolgenden Sprachsegments auslösen kann.The line-specific multiplexers M simultaneously emit a digital signal E to the selection circuit ZME when the last speech element generated in the filter set concerned is switched on, which digital signal identifies the completion of the speech synthesis process in the filter set. Like the line selection multiplexer ZMF, the selection circuit ZME brought from the current speech element selection signal SEA into a corresponding switching position now switches the corresponding digital signal through to the control unit StE, which can thus trigger the synthesis process of the next speech segment.

Abweichend von der Darstellung in FIG 2 können die Einzel-Abweichend von der Darstellung in FIG 2 können die Einzelfilter mit festen Koeffizienten auch einzeln adressierbar sein. In diesem Fall weist die Filterschaltung (F in FIG 2) keinen Zeilenauswahlmultiplexer (ZMF in FIG 2) und keine Auswahlchaltung (ZME in FIG 2) sowie keine zeilenindi- viduellen Multiplexer (M in FIG 2) der oben beschriebenen Art auf. Vielmehr verfügt die Steuereinheit (StE in FIG 2) dann über Einrichtungen zur Speicherung individuell adressierbarer Parameterwerte für die Anregungssignale und zur Zusamenschaltung der in den Einzelfiltern. erzeugbaren Sprachelementesignale. Der Filterwechseltaktgenerator FwG (in FIG 2) und die steuerbare Anregungs- Generatoreinrichtung G (in FIG 2) (ggf. mit Zeitfenster- schaltung) erfüllen auch bei dieser Ausführungsforn die oben beschriebenen Funktionen. Eine solche durch wahlfreie Ansteuerung von Einzelfiltern mittels einzelfilterindividueller Adressierung vidueller Adressierung gekenn zei chnete Aus führungsform braucht nur unterschiedliche Einzelfilter.. aufzuweisen, während bei dem Ausführungsbeispiel gemäß FIG 2 auch untereinander gleiche Einzalfilter in den verschiedenen Filtersätzen, ggf. auch in ein- und demselben Filtersatz angeordnet sein können. Die letzteren Alterna- tiven, die sich wegen der filtersatzindividuellen Adressierung gegenüber der einzelfilterindividuellen Adressierung mit geringerem steuertechnischen realisieren lassen, sind besonders für die wieder-Aufwand realisieren lassen, sind besonders für die wiedergabe von Sprache geeignet, die wiederholt gleiche Sprachsegmente enthält. Praktisch bedeutsam werden Ausführungsformen sein, die sowohl in Filtersätzen verbundene Einzelfilter als auch unabhängige Einzelfilter enthalten. Auf diese Weise läßt sich die Zahl der verwendeten Einzelfilter und der notwendige steuertechnische Aufwand optimieren.2, the individual filters can also be individually addressable with fixed coefficients. In this case, the filter circuit (F in FIG 2) no line selection multiplexer (ZMF in FIG. 2) and no selection circuit (ZME in FIG. 2) and no line-individual multiplexers (M in FIG. 2) of the type described above. Rather, the control unit (StE in FIG. 2) then has devices for storing individually addressable parameter values for the excitation signals and for interconnecting those in the individual filters. producible speech element signals. The filter change clock generator FwG (in FIG. 2) and the controllable excitation generator device G (in FIG. 2) (possibly with a time window circuit) also perform the functions described above in this embodiment. Such an embodiment, which is characterized by optional control of individual filters by means of individual filter-specific addressing vidual addressing, need only have different individual filters .., while in the exemplary embodiment according to FIG. 2 the same individual filters are also used in the different filter sets, possibly also in one and the same filter set can be arranged. The latter alternatives, which due to the addressing of the individual filter set compared to the addressing of the individual filter can be implemented with less control technology, are particularly suitable for the re-outlay, and are particularly suitable for the reproduction of language which repeatedly contains the same language segments. Embodiments that contain both individual filters connected in filter sets and independent individual filters will be of practical importance. In this way, the number of individual filters used and the necessary tax expenditure can be optimized.

Die erfindungsgemäß verwend eten Einzelfilter können auch als Linearpräidiktionsfilter mit festen n Koeffizienten gemäß FIG 3 ausgebildet sein. Lineare Prädikton als solche ist bekannt und in der einschlägigen Fachliteratur (so beispielsweise FLANAGAN, Speech Analysis Synthesis and Perception, Springer-Verlag Berlin, Heidelberg,The individual filters used according to the invention can also be designed as linear prediction filters with fixed n coefficients according to FIG. 3. Linear prediction tone as such is known and in the relevant specialist literature (for example FLANAGAN, Speech Analysis Synthesis and Perception, Springer-Verlag Berlin, Heidelberg,

New York 1972, S. 367 ff., S.390 ff.) beschrieben, so daß sich hier eine nähere Darstellung erübrigt. Dabei ist die erreichbare Sprachqualität in bestimmten Grenzen proportional zur Zahl der Koeffizienten.. Gute Sprachqualität läßt sich bereits mit etwa 10 Filter- koeffizienten realisieren. Ein solches Prädiktionsfilter wird mit seinen in FIG 3 mit A11 und B11 bezeichneten Anschlußklemmen an die entsprechend bezeichneten Anschlußklemmen der Filterschaltung F gemäß FIG 2 angeschlossen, um dort das Einzelfilter F11 zu bilden. Die erfindungs- gemäß verwendeten Linearpräidiktionsfilter können als Analog- oder als Digitalfilter ausgebildet sein. Dementsprechend werden den Filtern Anregungssignale in analoger oder digitaler Form vom Anregungsgenerator zugeführt; entsprechend ergeben sich an den Filterausgängen analoge oder digitale Signale.New York 1972, pp. 367 ff., P. 390 ff.), So that a detailed description is not necessary here. The speech quality that can be achieved is proportional to the number of coefficients within certain limits. Good speech quality can be achieved with around 10 filter coefficients. Such a prediction filter is connected with its connection terminals designated in FIG. 3 with A11 and B11 to the correspondingly designated connection terminals of the filter circuit F in accordance with FIG. 2 in order to form the individual filter F11 there. The linear prediction filters used according to the invention can be designed as analog or digital filters. Accordingly, the filters are supplied with excitation signals in analog or digital form from the excitation generator; accordingly, analog or digital signals result at the filter outputs.

Die erfindungsgemäß verwendeten Einzelfilter können, wie FIG 4 dargestellt auch als Formantfilter mit festen Filterkoeffizienter ausgebildet sein, wobei jedem Einzelfilter eine Parallelschaltung von drei Formantfiltern Nachbildung mindestens der ersten drei (niederfrequenten) Sprachfomanten entsprechen kann. Die Spracherzeugung durch Formantsynthese ist bekannt und hinreichend beschrieben (so beispielsweise in der bereits zitierten Literatur FLANAGAN S. 339 ff.), so daß sich eine Darstellung an dieser Stelle wiederum erübrigt. Die Formantfilter sind zweckmäßigerweise als Bandpässe mit bestimmten festen Durchlaßbereichen und Mittenfrequenzen dieser Durchlaßbereiche ausgebildet. Auch eine solche Filterschaltung kann in Analog- und Digitaltechnik realisiert sein.The individual filters used according to the invention can, as shown in FIG. 4, also be designed as formant filters with fixed filter coefficients, wherein each individual filter can correspond to a parallel connection of three formant filters simulating at least the first three (low-frequency) speech novels. The generation of speech by formant synthesis is known and adequately described (for example in the literature FLANAGAN p. 339 ff. Already cited), so that there is no need to describe it here. The formant filters are expediently designed as bandpasses with certain fixed pass bands and center frequencies of these pass bands. Such a filter circuit can also be implemented in analog and digital technology.

Die Einzelfilter können in allen vorstehend genannten Fällen in sogenannter CCD-Technik realisiert sein.The individual filters can be implemented in so-called CCD technology in all of the above cases.

Bei Verwendung von Transversalfiltern beziehungsweise Rekursivfiltern wird das Anregungssignal den Einzelfiltern in zeitdiskreter Form zugeführt. Hierzu kann die Filterschaltung (F in FIG 2) eine in FIG 2 nicht näher dargestellte Zeitfensterschaltung enthalten. Die Zeitfensterschaltung kann entsprechend dem Abtasttheorem ein Abtastsignal fester Frequenz erzeugen, die mindestens die doppelte Frequenz bezogen auf das abzutastende Netzsignal aufweist. Mit diesem so erzeugten Abtastsignal werden die steuerbare Anregungsgeneratoreinrichtung G sowie sämtliche Einzelfilter in die Filterschaltung F getaktet.When using transversal filters or recursive filters, the excitation signal is fed to the individual filters in a time-discrete form. For this purpose, the filter circuit (F in FIG. 2) can contain a time window circuit not shown in FIG. 2. According to the sampling theorem, the time window circuit can generate a sampling signal of fixed frequency, which has at least twice the frequency with respect to the network signal to be sampled. The controllable excitation generator device G and all individual filters are clocked into the filter circuit F with this scanning signal generated in this way.

Claims (11)

1. Schaltungsanordnung zur elektronischen Sprachsynthese, wobei Sprachelemente durch signifikante Parameter dargestellt werden und einzelne Sprachelemente zu längeren Sprachsegmenten zusammengesetzt werden können, wobei ein Anregungssignal zur Darstellung stimmhafter bzw. stimmloser Laute durch einen Ton- bzw. Rauschgenerator unter Verwendung mindestens eines Teils der signifikanten Parameter erzeugt und einer Filterschaltung zugeführt wird und wobei die filterausgangsseitigen elektrischen Signale zur akustischen Wiedergabe der gewünschten Sprachelemente und -segmente aufbereitet werden; dadurch gekennzeichnet, daß die Filterschaltung aus Einzelfiltern besteht, von denen jeweils nur diejenigen Einzelfilter, deren Filterkoeffizienten die restlichen signifikanten Parameter der tatsächlich sequentiell gewünschten Sprachelemente darstellen, sequentiell angesteuert werden.1. Circuit arrangement for electronic speech synthesis, speech elements being represented by significant parameters and individual speech elements being able to be assembled into longer speech segments, an excitation signal for representing voiced or unvoiced sounds being generated by a tone or noise generator using at least some of the significant parameters and is fed to a filter circuit and the filter output-side electrical signals are processed for acoustic reproduction of the desired speech elements and segments; characterized in that the filter circuit consists of individual filters, of which only those individual filters, the filter coefficients of which represent the remaining significant parameters of the speech elements actually desired sequentially, are sequentially controlled. 2. Schaltungsanordnung nach Anspruch 1, dadurch gekennzeichnet, daß in Analogtechnik aufgebaute Einzelfilter vorgesehen sind, die mit einem zeitdiskreten Analog-Anregungssignal beaufschlagbar sind.2. Circuit arrangement according to claim 1, characterized in that individual filters constructed in analog technology are provided which can be acted upon with a discrete-time analog excitation signal. 3. Schaltungsanordnung nach Anspruch 1, dadurch gekennzeichnet,daß in Digitaltechnik aufgebaute Einzelfilter vorgesehen sind, die mit einem zeitdiskreten Digital-Anregungssignal beaufschlagbar sind und daß das gewonnene Sprachsignal zur sprachlichen Wiedergabe digital/analoggewandelt wird.3. Circuit arrangement according to claim 1, characterized in that individual filters constructed in digital technology are provided which can be acted upon with a discrete-time digital excitation signal and that the speech signal obtained is converted digitally / analogously for speech reproduction. 4. Schaltungsanordnung nach Anspruch 2 oder 3, gekennzeichnet durch wahlfreie Ansteuerung der Einzelfilter durch einzelfilterindividuelle Adressierung.4. Circuit arrangement according to claim 2 or 3, characterized by optional control of the individual filter by individual filter addressing. 5. Schaltungsanordnung nach Anspruch 2 oder 3, dadurch gekennzeichnet, βdie Einzelfilter in Filtersätzen zur Darstellung längerer Sprachsegmente angeordnet sind, wobei eine wahlfreie Ansteuerung der filtersätze durch filtersatzindividuelle Adressierung erfolgt.5. Circuit arrangement according to claim 2 or 3, characterized in that β the individual filters are arranged in filter sets for displaying longer language segments, an optional control of the filter sets takes place by addressing individual filter sets. 6. Schaltungsanordnung nach Anspruch 5, dadurch gekennzeichnet, daß die Einzelfilter in einer Matrix angeordnet sind, in der jeweils die Einzelfilter einer Mätrixzeile parallel mit dem jeweiligen Anregungssignal beaufschlagt werden und die Einzelfilterausgänge jeweils einer Matrixzeile sequentiell mit dem Matrixausgang verbindbar sind.6. Circuit arrangement according to claim 5, characterized in that the individual filters are arranged in a matrix, in each of which the individual filters of a matrix line are acted upon in parallel with the respective excitation signal and the individual filter outputs of one matrix line can be connected sequentially to the matrix output. 7. Schaltungsanordnung nach Anspruch 5 oder 6, dadurch gekennzeichnet, daß pro Filtersatz eine Speichereinheit vorgesehen ist, in der Parameter des Anregungssignals zur Darstellung eines Sprachsegments abgespeichert sind.7. Circuit arrangement according to claim 5 or 6, characterized in that a storage unit is provided per filter set, in which parameters of the excitation signal for representing a speech segment are stored. 8. Schaltungsanordnung nach einem der Ansprüche 4 bis 7, dadurch gekennzeichnet, daß die Einzelfilter als sogenannte Linear-Predictive-Filter ausgebildet sind.8. Circuit arrangement according to one of claims 4 to 7, characterized in that the individual filters are designed as so-called linear predictive filters. 9 . Schaltungsanordnung nach einem der Ansprüche 4 bis 7 dadurch gekennzeichnet, daβ die Einzelfilter als Formantfilter mit festen Formantmittenfrequenz- und -bandbreitenkoeffizienten zur Erzeugung von Sprache durch die Wiedergabe mindestens der drei tiefsten Formanten ausgebildet sind.9. Circuit arrangement according to one of Claims 4 to 7, characterized in that the individual filters are designed as formant filters with fixed formant center frequency and bandwidth coefficients for generating speech by reproducing at least the three lowest formants. 10 . Schaltungsanordnung nach einem der Ansprüche 4 bis 9, dadurch gekennzeichnet, daß die Einzelfilter in sogenannter CCD-Technik realisiert sind.10th Circuit arrangement according to one of claims 4 to 9, characterized in that the individual filters are implemented in so-called CCD technology. 11. Schaltungsanordnung nach einem der Ansprüche 4 bis 10 , dadurchgekennzeichnet, daß die Einzelfilter als Transversalfilter ausgebildet sind.11. Circuit arrangement according to one of claims 4 to 10, characterized in that the individual filters are designed as transversal filters.
EP83104873A 1982-05-18 1983-05-17 Arrangement for electronic speech synthesis Expired EP0094681B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AT83104873T ATE26354T1 (en) 1982-05-18 1983-05-17 CIRCUIT ARRANGEMENT FOR ELECTRONIC SPEECH SYNTHESIS.

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE3218755 1982-05-18
DE3218755A DE3218755A1 (en) 1982-05-18 1982-05-18 CIRCUIT ARRANGEMENT FOR THE ELECTRONIC VOICE SYNTHESIS

Publications (2)

Publication Number Publication Date
EP0094681A1 true EP0094681A1 (en) 1983-11-23
EP0094681B1 EP0094681B1 (en) 1987-04-01

Family

ID=6163962

Family Applications (1)

Application Number Title Priority Date Filing Date
EP83104873A Expired EP0094681B1 (en) 1982-05-18 1983-05-17 Arrangement for electronic speech synthesis

Country Status (5)

Country Link
US (1) US4694496A (en)
EP (1) EP0094681B1 (en)
JP (1) JPS58205200A (en)
AT (1) ATE26354T1 (en)
DE (2) DE3218755A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0311022A2 (en) * 1987-10-06 1989-04-12 Kabushiki Kaisha Toshiba Speech recognition apparatus and method thereof

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19860133C2 (en) * 1998-12-17 2001-11-22 Cortologic Ag Method and device for speech compression

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0037552A1 (en) * 1980-04-03 1981-10-14 Kabushiki Kaisha Toshiba Signal synthesizer apparatus

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US2060321A (en) * 1936-02-18 1936-11-10 Jr Elmer E Johnson Safety razor
US2121142A (en) * 1937-04-07 1938-06-21 Bell Telephone Labor Inc System for the artificial production of vocal or other sounds
US2194298A (en) * 1937-12-23 1940-03-19 Bell Telephone Labor Inc System for the artificial production of vocal or other sounds
US2881257A (en) * 1956-08-16 1959-04-07 Bell Telephone Labor Inc Spectrum synthesizer
US3624301A (en) * 1970-04-15 1971-11-30 Magnavox Co Speech synthesizer utilizing stored phonemes
US3836717A (en) * 1971-03-01 1974-09-17 Scitronix Corp Speech synthesizer responsive to a digital command input
NL7306902A (en) * 1972-05-26 1973-11-28
NL7902238A (en) * 1978-04-27 1979-10-30 Kawai Musical Instr Mfg Co DEVICE FOR GENERATING A VOCAL SOUND SIGNAL IN AN ELECTRONIC MUSICAL INSTRUMENT.
US4454609A (en) * 1981-10-05 1984-06-12 Signatron, Inc. Speech intelligibility enhancement
US4475228A (en) * 1981-11-27 1984-10-02 Bally Manufacturing Corporation Programmable sound circuit for electronic games

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0037552A1 (en) * 1980-04-03 1981-10-14 Kabushiki Kaisha Toshiba Signal synthesizer apparatus

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
ELEKTOR, Band 7, Nr. 5, Mai 1981, Seiten 17-19, Canterbury, GB *
TELECOMMUNICATIONS AND RADIO ENGINEERING, Band 25/26, Nr. 3, März 1971, Seiten 4-13, Washington, US *
THE JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA, Band 66, Nr. 5, November 1979, Seiten 1325-1332, New York, US *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0311022A2 (en) * 1987-10-06 1989-04-12 Kabushiki Kaisha Toshiba Speech recognition apparatus and method thereof
EP0311022A3 (en) * 1987-10-06 1990-02-28 Kabushiki Kaisha Toshiba Speech recognition apparatus and method thereof
US5001760A (en) * 1987-10-06 1991-03-19 Kabushiki Kaisha Toshiba Speech recognition apparatus and method utilizing an orthogonalized dictionary

Also Published As

Publication number Publication date
DE3370707D1 (en) 1987-05-07
JPS58205200A (en) 1983-11-30
DE3218755A1 (en) 1983-11-24
US4694496A (en) 1987-09-15
EP0094681B1 (en) 1987-04-01
ATE26354T1 (en) 1987-04-15

Similar Documents

Publication Publication Date Title
DE68919637T2 (en) Method and device for speech synthesis by covering and summing waveforms.
DE69816221T2 (en) LANGUAGE SPEED CHANGE METHOD AND DEVICE
DE2115258A1 (en) Speech synthesis by concatenating words encoded in formant form
DE2229149A1 (en) Method of transmitting speech
DE3590157T (en) System for storing and retrieving the voice of a telephone operator
DE2920298A1 (en) BINARY INTERPOLATOR CIRCUIT FOR AN ELECTRONIC MUSICAL INSTRUMENT
WO1983000231A1 (en) Method for testing analog/digital converters and/or digital/analog converters or sections of information transmission circuits, comprising such converters or connected in series therewith, for communication apparatus, particularly for testing coders-decoders for pcm apparatus, as well as device for implemeting such method
EP0126962A2 (en) Electronic keyboard musical instrument, and device for using it
DE2622423B2 (en) Electrical arrangement for the transmission or storage of a speech or sound signal in coded form
DE19720651A1 (en) Digital hearing aid
DE3006339C2 (en) Speech synthesizer
DE2836738C3 (en) Method for recording sound signals produced when playing a key-operated musical instrument and the associated playing expressions, in particular the volume, corresponding data signal sequences on a recording medium and devices for carrying out the method
DE69717377T2 (en) Sprachgeschwindigkeitsumwandler
DE2836736C3 (en) Method for recording sound signals produced when playing a key-operated musical instrument and the associated playing expressions, in particular the volume, corresponding data signal sequences on a recording medium and device for carrying out the method
EP0094681B1 (en) Arrangement for electronic speech synthesis
DE2836737C3 (en) Method for recording on a recording medium data signal sequences corresponding to sound signals produced when playing a key-operated musical instrument
EP0126975A2 (en) Electronic keyboard musical instrument
DE3246712C2 (en)
DE2649540A1 (en) Speech synthesis system using time quantised signals - has discrete sets of amplitudes and phases Fourier transform processed
DE2854601C2 (en) Process for sound processing and sound synthesizer for its implementation
DE3037276A1 (en) TONSYNTHESIZER
DE2826570C2 (en)
DE3232835C2 (en)
DE2431989A1 (en) PROCESS AND EQUIPMENT FOR GENERATING ARTIFICIAL REVIEWS
DE2209548C3 (en) Electric speech synthesizer circuit

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

Designated state(s): AT DE FR GB

17P Request for examination filed

Effective date: 19840521

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AT DE FR GB

REF Corresponds to:

Ref document number: 26354

Country of ref document: AT

Date of ref document: 19870415

Kind code of ref document: T

REF Corresponds to:

Ref document number: 3370707

Country of ref document: DE

Date of ref document: 19870507

ET Fr: translation filed
PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

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

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

26N No opposition filed
PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GB

Effective date: 19890517

Ref country code: AT

Effective date: 19890517

GBPC Gb: european patent ceased through non-payment of renewal fee
PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: FR

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 19900131

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DE

Effective date: 19900201

REG Reference to a national code

Ref country code: FR

Ref legal event code: ST