-
公开(公告)号：JP2001296880A
公开(公告)日：2001-10-26
申请号：JP2001084632
申请日：2001-03-23
Applicant: LUCENT TECHNOLOGIES INC
Inventor： KIRAZ GEORGE A , OLIVE JOSEPH PHILIP , SHIH CHI-LIN
Abstract: PROBLEM TO BE SOLVED: To identify plural plausible pronunciations of a given person's name and to apply a set of such 'allowable' pronunciations to a specific speaker group. SOLUTION: The method is used to generate plural plausible pronunciations of an intrinsic name, i.e., the method is used to execute voice recognition of uttering including a person's intrinsic name within a given speaker group. The method has (a) a step which identifies more than one language among plural languages as the origins of a possible intrinsic name and (b) a step which generates plural plausible pronunciations for the given intrinsic name based on identified languages and more than one characteristic related to the given speaker group. The characteristic of the group is for example a national origin of the group (the mother language speaker of the origin language of the intrinsic name has a high probability of using a character-sound conversion rule of the mother language).
-
公开(公告)号：CA2222582A1
公开(公告)日：1997-02-27
申请号：CA2222582
申请日：1996-08-02
Applicant: LUCENT TECHNOLOGIES INC
Abstract: A speech synthesis method employs an acoustic element database that is established from phonetic sequences occurring in an interval of a speech signal in establishing the database, trajectories are determined (220) for each of the phonetic sequences containing a phonetic segment that corresponds to a particular phoneme (210). A tolerance region is then identified based on a concentration of trajectories that correspond to different phoneme sequences (230). The acoustic elements for the database (260) are formed from portions of the phonetic sequences by identifying cut points (250) in the phonetic sequences which corespond to time points along the respective trajectories proximate the tolerance region (240). In this manner, it is possible to concatenate the acoustic elements having a common junction phonemes such that perceptible discontinuities at the junction phonemes are minimized. Computationally simple and fast methods for determining the tolerance region are also disclosed.
-
公开(公告)号：DE60000138T2
公开(公告)日：2002-10-31
申请号：DE60000138
申请日：2000-10-23
Applicant: LUCENT TECHNOLOGIES INC
Inventor： KIRAZ GEORGE A , OLIVE JOSEPH PHILIP , SHIH CHI-LIN
Abstract: Multiple, yet plausible, pronunciations of a proper name are generated based on one or more potential language origins of the name, and based further on the context in which the name is being spoken -- namely, on characteristics of the population of potential speakers. Conventional techniques may be employed to identify likely candidates for the language origin of the name, and the characteristics of the speaker population on which the generation of the pronunciations is further based may comprise, for example, the national origin of the speakers, the purpose of the speech, the geographical location of the speakers, or the general level of sophistication of the speaker population. Specifically, a method and apparatus is provided for generating a plurality of plausible pronunciations for a proper name, the method or apparatus for use in performing speech recognition of speech utterances comprising the proper name by individuals within a given population of speakers, the method or apparatus comprising steps or means respectively for (a) identifying one or more of a plurality of languages as a potential origin of the proper name; and (b) generating a plurality of plausible pronunciations for the given proper name, one or more of the plurality of pronunciations based on the one or more identified languages, and the plurality of plausible pronunciations based further on one or more characteristics associated with the given population of speakers.
-
公开(公告)号：CA2222582C
公开(公告)日：2001-09-11
申请号：CA2222582
申请日：1996-08-02
Applicant: LUCENT TECHNOLOGIES INC
Abstract: A speech synthesis method employs an acoustic element database that is established from phonetic sequences occurring in an interval of a speech signal in establishing the database, trajectories are determined (220) for each of the phonetic sequences containing a phonetic segment that correspond s to a particular phoneme (210). A tolerance region is then identified based o n a concentration of trajectories that correspond to different phoneme sequenc es (230). The acoustic elements for the database (260) are formed from portions of the phonetic sequences by identifying cut points (250) in the phonetic sequences which corespond to time points along the respective trajectories proximate the tolerance region (240). In this manner, it is possible to concatenate the acoustic elements having a common junction phonemes such tha t perceptible discontinuities at the junction phonemes are minimized. Computationally simple and fast methods for determining the tolerance region are also disclosed.
-
公开(公告)号：DE60000138D1
公开(公告)日：2002-05-29
申请号：DE60000138
申请日：2000-10-23
Applicant: LUCENT TECHNOLOGIES INC
Inventor： KIRAZ GEORGE A , OLIVE JOSEPH PHILIP , SHIH CHI-LIN
Abstract: Multiple, yet plausible, pronunciations of a proper name are generated based on one or more potential language origins of the name, and based further on the context in which the name is being spoken -- namely, on characteristics of the population of potential speakers. Conventional techniques may be employed to identify likely candidates for the language origin of the name, and the characteristics of the speaker population on which the generation of the pronunciations is further based may comprise, for example, the national origin of the speakers, the purpose of the speech, the geographical location of the speakers, or the general level of sophistication of the speaker population. Specifically, a method and apparatus is provided for generating a plurality of plausible pronunciations for a proper name, the method or apparatus for use in performing speech recognition of speech utterances comprising the proper name by individuals within a given population of speakers, the method or apparatus comprising steps or means respectively for (a) identifying one or more of a plurality of languages as a potential origin of the proper name; and (b) generating a plurality of plausible pronunciations for the given proper name, one or more of the plurality of pronunciations based on the one or more identified languages, and the plurality of plausible pronunciations based further on one or more characteristics associated with the given population of speakers.
-
Patent Agency Ranking
公开(公告)号：BR9612624A
公开(公告)日：2000-05-23
申请号：BR9612624
申请日：1996-08-02
Applicant: LUCENT TECHNOLOGIES INC
Abstract: A speech synthesis method employs an acoustic element database that is established from phonetic sequences occurring in an interval of a speech signal. In establishing the database, trajectories are determined for each of the phonetic sequences containing a phonetic segment that corresponds to a particular phoneme. A tolerance region is then identified based on a concentration of trajectories that correspond to different phoneme sequences. The acoustic elements for the database are formed from portions of the phonetic sequences by identifying cut points in the phonetic sequences which correspond to time points along the respective trajectories proximate the tolerance region. In this manner, it is possible to concatenate the acoustic elements having a common junction phonemes such that perceptible discontinuities at the junction phonemes are minimized. Computationally simple and fast methods for determining the tolerance region are also disclosed.
-
公开(公告)号：CA2336459A1
公开(公告)日：2001-09-27
申请号：CA2336459
申请日：2001-02-14
Applicant: LUCENT TECHNOLOGIES INC
Inventor： SHIH CHI-LIN , OLIVE JOSEPH PHILIP , KIRAZ GEORGE A
Abstract: Multiple, yet plausible, pronunciations of a proper name are generated based on one or more potential language origins of the name, and based further on the context in which the name is being spoken -- namely, on characteristics of the populati on of potential speakers. Conventional techniques may be employed to identify like ly candidates for the language origin of the name, and the characteristics of t he speaker population on which the generation of the pronunciations is further based ma y comprise, for example, the national origin of the speakers, the purpose of the speech, the geographical location of the speakers, or the general level of sophisticatio n of the speaker population. Specifically, a method and apparatus is provided for generating a plurality of plausible pronunciations for a proper name, the method or apparatus for use in performing speech recognition of speech utterances comprising the proper nam e by individuals within a given population of speakers, the method or apparatus comprising steps or means respectively for (a) identifying one or more of a plurality o f languages as a potential origin of the proper name; and (b) generating a plurality of plausible pronunciations for the given proper name, one or more of the plurality of pronunciations based on the one or more identifiedlanguages, and the plurality of plausible pronunciations based further on one or more characteristics associated with the given population of speakers.
-
公开(公告)号：MX9801086A
公开(公告)日：1998-04-30
申请号：MX9801086
申请日：1996-08-02
Applicant: LUCENT TECHNOLOGIES INC
Abstract: La presente invencion se refiere a un método para síntesis de habla que emplea una base de datos de elementos acusticos, que se establece a partir de secuencias fonéticas que ocurren en un intervalo de una señal de habla, al establecer la base de datos, se determinan trayectorias por cada una de las secuencias fonéticas que contienen un segmento fonético que corresponde a un fonema particular. Luego se identifica una region de tolerancia con base en una concentracion de trayectorias que corresponden a diferentes secuencias de fonema. Los elementos acusticos para la base de datos se forman a partir de porciones de las secuencias fonéticas, al identificar puntos de corte en las secuencias fonéticas que corresponden a puntos en tiempo sobre las trayectorias respectivas proximas a la region de tolerancia. De esta manera, es posible concatenar los elementos acusticos que tienen fonemas de union comunes tal que las discontinuidades perceptibles en los fonemas de union se minimizan. También se describen métodos computacionalmente simples y rápidos para determinar la region de tolerancia.
-
公开(公告)号：AU6645096A
公开(公告)日：1997-03-12
申请号：AU6645096
申请日：1996-08-02
Applicant: LUCENT TECHNOLOGIES INC
Abstract: A speech synthesis method employs an acoustic element database that is established from phonetic sequences occurring in an interval of a speech signal. In establishing the database, trajectories are determined for each of the phonetic sequences containing a phonetic segment that corresponds to a particular phoneme. A tolerance region is then identified based on a concentration of trajectories that correspond to different phoneme sequences. The acoustic elements for the database are formed from portions of the phonetic sequences by identifying cut points in the phonetic sequences which correspond to time points along the respective trajectories proximate the tolerance region. In this manner, it is possible to concatenate the acoustic elements having a common junction phonemes such that perceptible discontinuities at the junction phonemes are minimized. Computationally simple and fast methods for determining the tolerance region are also disclosed.
-
公开(公告)号：EP0845139A4
公开(公告)日：1999-10-20
申请号：EP96926228
申请日：1996-08-02
Applicant: LUCENT TECHNOLOGIES INC
CPC classification number: G10L13/02
Abstract: A speech synthesis method employs an acoustic element database that is established from phonetic sequences occurring in an interval of a speech signal. In establishing the database, trajectories are determined for each of the phonetic sequences containing a phonetic segment that corresponds to a particular phoneme. A tolerance region is then identified based on a concentration of trajectories that correspond to different phoneme sequences. The acoustic elements for the database are formed from portions of the phonetic sequences by identifying cut points in the phonetic sequences which correspond to time points along the respective trajectories proximate the tolerance region. In this manner, it is possible to concatenate the acoustic elements having a common junction phonemes such that perceptible discontinuities at the junction phonemes are minimized. Computationally simple and fast methods for determining the tolerance region are also disclosed.
-
-
-
-
-
-
-
-
-
Search Results
10
records
(took 0.022 seconds)
