WO2010148893A1 - Method and system for unified multi-language development and voice play in media server - Google Patents

Method and system for unified multi-language development and voice play in media server Download PDF

Info

Publication number
WO2010148893A1
WO2010148893A1 PCT/CN2010/073282 CN2010073282W WO2010148893A1 WO 2010148893 A1 WO2010148893 A1 WO 2010148893A1 CN 2010073282 W CN2010073282 W CN 2010073282W WO 2010148893 A1 WO2010148893 A1 WO 2010148893A1
Authority
WO
WIPO (PCT)
Prior art keywords
words
control
internal
language
rules
Prior art date
Application number
PCT/CN2010/073282
Other languages
French (fr)
Chinese (zh)
Inventor
胡锡文
蒋恒
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2010148893A1 publication Critical patent/WO2010148893A1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L65/00Network arrangements, protocols or services for supporting real-time applications in data packet communication
    • H04L65/10Architectures or entities
    • H04L65/1063Application servers providing network services
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L13/00Speech synthesis; Text to speech systems
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L65/00Network arrangements, protocols or services for supporting real-time applications in data packet communication
    • H04L65/10Architectures or entities
    • H04L65/102Gateways
    • H04L65/1043Gateway controllers, e.g. media gateway control protocol [MGCP] controllers
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04MTELEPHONIC COMMUNICATION
    • H04M2203/00Aspects of automatic or semi-automatic exchanges
    • H04M2203/35Aspects of automatic or semi-automatic exchanges related to information services provided via a voice call
    • H04M2203/355Interactive dialogue design tools, features or methods
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04MTELEPHONIC COMMUNICATION
    • H04M3/00Automatic or semi-automatic exchanges
    • H04M3/42Systems providing special services or facilities to subscribers
    • H04M3/487Arrangements for providing information services, e.g. recorded voice services or time announcements
    • H04M3/493Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
    • H04M3/4936Speech interaction details

Definitions

  • G is provided with each of the required resources under the exchange or under the control of each. Looking at the large scales that are not the same as G, each of them is indispensable, and the main function of the function is more important, and the different functions provide different functions.
  • the sound flow of each of them is exchanged by the voice signaling, so that the list of good speeches and broadcasts, and the received sound signaling lines are parsed, and the sound, sound content, sound, etc. are extracted.
  • the rules of law are combined into a final list. The main thing is that Hugh is not able to support all the different or different rules of the law, and the new words are related to the version of the more, not costly and unfavorable.
  • the wood to be solved by the Institute is to provide a method and system for multi-language and sound, and to support the existing resources at a lower cost.
  • this method provides a method for multi-speech and sound.
  • Hume is divided according to the words, and the internal words of the words are only in the control of the exchange of the rules of speech and the internal words of the various internal words.
  • Each of the control commands is analyzed, and the internal control words of the control commands in the control command are obtained, and the legal rules of the internal words of the internal control words are analyzed, and the audio information lines of the legal rules are analyzed and combined into new ones. list of.
  • Steps are not said, the method is near the rules of the law
  • the audio information lines are parsed into a new list, and the new list sound is overtoned.
  • Steps are not said, the method is close to each internal statement, and the list of legal rule combinations of the internal words of the rest is sequentially sequenced in multiple words.
  • this book provides a system for the simultaneous promotion of friends and voices.
  • Steps are not said, the above-mentioned squatting method will be based on the words and demerits, and will introduce different dialects or interpreters with less dialects.
  • Steps are not said, the control mode of the underlying mode exchange system and the internal words of the words of the internal words of the Hughes are described.
  • Steps are not said, if the control in the externally parsed control command has multiple internal statements, then the list of legal rule combinations of the internal speeches is sequentially sequenced.
  • the content of the project is that, in the various places, according to the words and deeds, the words of each of the internal words, the control of the words and the internal words of the rest
  • the different control of the words can go to the internal words or words.
  • the internal rules of the law rule can be parsed and combined in the same legal rules.
  • the user only needs to specify the list, and the content of the multicast and broadcast needs.
  • Xiu is in Shanghai, and we can introduce the general and Shanghai scores. , ordinary and Shanghai can have the same rest and the same rules of law.
  • Step 202 Controlling the exchange system and the internal words of the rest
  • Step 203 Exchanging the control command specified by the command, and controlling the voice information only in the control command
  • Step 204 Each of the control commands is parsed, and the internal control words of the control statements in the control command are obtained, the legal rules of the spoken words of the internal words, and the audio information lines of the legal rules are analyzed and combined. New list
  • Step 205 resolves the list of audio tones.
  • step 201 the description in step 201 will be followed. Divided into words,
  • the words can be considered more than the words with the same rules of law, or the words of the geographical phase, etc., Hugh’s method of dichotomy can be considered, as long as the method can this.
  • Hugh analyzes the words in the control, and then combines the internal words, then combines the list and broadcasts in multiple languages. Hugh can be used to control the number of sounds and the number of sounds in a sequence.
  • the words that need to be exchanged are parsed into different dialects, and the typical ones in the typical broadcasts are first broadcasted and then re-broadcasted in other dialects, and the words can be put into the internal words.
  • the list of the internal words in advance is in advance, and the rules of the law are obtained.
  • Information line parsing and combination In many internal statements, only the list in the parsing is different, and the parsing rules used are the same rules.
  • Step 1 in the various devices, will be in accordance with the words, rest each internal words.
  • a dialect with similar or similar dialects can be used.
  • step 2 in the control of the exchange system, the words in the control system and the words in the internal control are configured in the local dialect.
  • step 3 the exchange control system specifies the words, the number of sounds, and the sound content specified in the order.
  • step 4 Hugh receives the control command, controls the analysis control command, and analyzes the speech, the number of times, and the content of the sound.
  • Step 5 the control words of each configuration and the internal words of the rest, you need the internal words of the tone.
  • step 6 the legal rules of each of the phases are combined into a new list, and the list and the sound of the tone are parsed.
  • Step 7 If step 5 resolves the words in the control, each internal multi-word, in step 6, the forest combines the list, and the multi-sequence broadcasts in sequence.
  • Step 2 is close, and if it is in the home of the sheep, it is not necessary to configure the internal words of the control. Hugh is still required to broadcast according to the original good sound, and can be configured.
  • Switching system 31 in the control order specified by the control, in the control order, control words and audio information
  • Step S2 in accordance with the dialect, the dialect and the dialect most likely to be broadcast. In Hugh, the most likely words to broadcast
  • Step S3 configured in the rest device, step S2, in the control of the exchange system of the dialect
  • step S4 the switching system controls the broadcaster to specify the support of the new dialect.
  • step S3 the dialect can be exchanged with the control of the exchange system. It can also be compared with the control of multiple exchange systems. In the broadcast, as long as the words established in any dialect are broadcast, the dialect can be broadcast.
  • step S4 the rests receive the voice signaling of the switching system, and the new basics of the broadcast are combined according to the proposed rule, and the sound is sounded.
  • the configuration of the control is not the same. In this step, you can specify the control word and any or more dialects in the configuration. In addition, the same rule, combined Chinese dialect and list are not combined, reducing the workload and the weight of each function.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Telephonic Communication Services (AREA)
  • Machine Translation (AREA)

Abstract

The present invention provides a method and system for unified multi-language development and voice play in a media server. The media server divides voices into language classes according to language types, sets media server internal language identifiers corresponding to language classes and sets a correspondence between language classes and grammar rules. Set a correspondence between language identifiers in control protocols of a soft switching system and media server internal language identifiers. The soft switching system issues a control signaling to play the voice, and the control signaling includes the control protocol language identifier and voice information. The media server parses the control signaling, obtains the media server internal language identifier corresponding to the control protocol language identifier in the control signaling, obtains the corresponding grammar rule according to the language class corresponding to the media server internal language identifier, and combines the voice information into a new voice element list according to the grammar rule after parsing the voice information. The present invention can reduce the development effort required for the development of new dialects.

Description

休 各器多倍言統 和 音的方法及 統 木領域  How to use multiple voices and sounds in various fields and the field of wood
本 涉及 G ( ex Ge e a o ewok, 下 代同 ), 休涉及 到在 休 各器中提供 支持多 言統 和 音的方法及 。 背景 木 This is related to G ( ex Ge e a o ewok, the next generation), and Hugh is involved in providing methods to support multi-lingual harmony in the various devices. Background
G 同 休 各 在 交換或者 各 的控制下, 提供各 各所需的 休資源。 看 各的不 和 G 同 的大規模 , 休 各 逐漸 各 中不可或缺的 各, 其中 通 功能更 力主要功能 , 何在不同 不同 提供 活的 功能提出了更高的要求。  G is provided with each of the required resources under the exchange or under the control of each. Looking at the large scales that are not the same as G, each of them is indispensable, and the main function of the function is more important, and the different functions provide different functions.
目前 休 各 的 音流程, 1 所示, 由 交換 統下 音信 令, 令中 好的 言矣 和 播 的 列表, 休 各 收到 音信令 行解析, 提取出 矣 、 音內容、 音 等, 然 的 法規則組合成最終的 列表。 主要的 是 休 各 不可能支持所有的 法規則 不同 或者不同 的 言, 而 且 新增 言都要涉及到版本的更 , 不 成本 而且不利 于 的 和 。  At present, the sound flow of each of them, as shown in 1 , is exchanged by the voice signaling, so that the list of good speeches and broadcasts, and the received sound signaling lines are parsed, and the sound, sound content, sound, etc. are extracted. The rules of law are combined into a final list. The main thing is that Hugh is not able to support all the different or different rules of the law, and the new words are related to the version of the more, not costly and unfavorable.
休未 , 目前 休 各 的主要控制 是 248, 在 交換 統 音信令中必須 方 好的 言矣 , 中文 言的 言矣 般是 "ch ", 英文 言的 言矣 祇是 "e "。 需要增 言 , 方都需要更 版本, 休 各器和 交換 統的 合性太 , 限制 了 休 各 的 。 休 各 作力 的 統 各, 在不同 , 因此就有不 同方言的 音要求, 在中 尤其突出的是普通 、 、 上海 等。 統的解決方案是新增 方言 按照新的 言未 , 除了和 交換 統 同 新的 言 外, 休 各 內部近 新增 法規則和 。 同 不同 的方言往往只是 音的不同, 在 法 規則上基本是 致的, 因此 模式不 增 了 成本, 而且增 了 統的 余 。 內容 Hugh, the main control of Xiu is 248. In the exchange of voice signaling, it must be a good statement. The Chinese language is "ch", and the English language is only "e". Need to add more, the party needs a more version, and the combination of the various devices and the exchange system is too limited, which limits the rest. The various functions of Hugh's efforts are different, so there are different phonetic requirements for dialects, especially in the ordinary, Shanghai, and so on. The solution is to add new dialects in accordance with the new words. In addition to the new words and exchanges, the internal rules of the law are added. Different dialects are often only different in sound, and basically in the rules of law, so the model does not increase the cost, but also increases the balance. content
本 所要解決的 木 是提供 休 各器多 言統 和 音的方法及 統, 以較低成本 新增 言的支持, 合理利用 休 各 現有的資源。  The wood to be solved by the Institute is to provide a method and system for multi-language and sound, and to support the existing resources at a lower cost.
了解決上述 木 , 本 提供了 休 各器多 言統 和 音的方法,  In order to solve the above-mentioned wood, this method provides a method for multi-speech and sound.
休 各 按照 言 分力 言 , 所述 言 相 的 休 各 內部 言 祇及 所述 言 法規則的 交換 統的控制 中 言 和所述 休 各 內部 言 的  Hume is divided according to the words, and the internal words of the words are only in the control of the exchange of the rules of speech and the internal words of the various internal words.
交換 統下 控制 令播 , 所述控制 令中 控制 言 祇及 音信息  Exchanging the control command, the control command, the control message, and the sound information
休 各 解析所述控制 令, 得 所述控制 令中的控制 言 相 的 休 各 內部 言 , 所述 休 各 內部 言 的 言 得 的 法規則, 所述 法規則 所述 音信息 行解析 組合成新的 列表。  Each of the control commands is analyzed, and the internal control words of the control commands in the control command are obtained, and the legal rules of the internal words of the internal control words are analyzed, and the audio information lines of the legal rules are analyzed and combined into new ones. list of.
步未說, 所述方法近 所述 休 各 所述 法規則 所述 音信息 行解析 組合成新的 列表 , 通 音 所述新 的 列表 音。 Steps are not said, the method is near the rules of the law The audio information lines are parsed into a new list, and the new list sound is overtoned.
步 , 所述 休 各 按照 言 分力 言 休 將 介 相 近 的多 方言或者具有較少方言的 介 的 言分力 介 言 。  Steps, said Hugh each according to the words and demerits will be introduced in a number of dialects or a dialect with less dialects.
步未說, 所述 交換 統的控制 中 言 和所述 休 各 內部 言 的 休  The steps are not said, the control system of the exchange system and the rest of the internal words of the rest
控制 下 的 言 介 言 相 的 休 各 內部 言 相 。  Under the control of the words and speeches of the various internal statements.
步未說, 所述方法近 多 休 各 內部 言 , 則 所述 休 各 內部 言 的 言 相 的 法規則組合 列表, 多 言順序播 。 了解決上述 木 , 本 近提供了 休 各器多 言統 升友和 音的 統,  Steps are not said, the method is close to each internal statement, and the list of legal rule combinations of the internal words of the rest is sequentially sequenced in multiple words. In order to solve the above-mentioned wood, this book provides a system for the simultaneous promotion of friends and voices.
, 于 休 各器中, 于將 按照 言 分力 言 , 所述 言 相 的 休 各 內部 言 祇及 所述 言 法規則的 交換 統的控制 中 言 和所述 休 各 內部 言 的  In the various devices, in accordance with the words of the words, the internal words of the words and the exchange of the rules of the rules of speech and the internal words of the various internal words
交換 統, 于下 控制 令播 , 所述控制 令中 控制 言 祇及 音信息  Switching the system, controlling the broadcast, and controlling the voice information in the control command
, 于 休 各 , 于解析所述控制 令, 得 所述 控制 令中的控制 言 相 的 休 各 內部 言 ,  , in Hugh, in the analysis of the control order, the control words in the control order, the internal words,
所述 休 各 內部 言 的 言 得 的 法規則, 所述 法規則 所述 音信息 行解析 組合成新的 列表。 步未說, The rules of the spoken words of the internal words are parsed and combined into a new list. Did not say,
、 步未說, 所述 矣 下方式將 按照 言矣 分力 言 矣 將 介 相 近 的多 方言或者具有較少方言的 介 言分力 介 言 矣。  Steps are not said, the above-mentioned squatting method will be based on the words and demerits, and will introduce different dialects or interpreters with less dialects.
步未說, 所述 矣 下方式 交換 統的控制 中 言 和所述 休 各 內部 言 的 控制 下 的 言 介 言 矣相 的 休 各 內部 言 相 。  Steps are not said, the control mode of the underlying mode exchange system and the internal words of the words of the internal words of the Hughes are described.
步未說, 若所述外 解析出的所述控制 令中的控制 言 多 休 各 內部 言 , 則 所述 休 各 內部 言 的 言 矣相 的 法規則組合 列表, 多 言順序播 。  Steps are not said, if the control in the externally parsed control command has multiple internal statements, then the list of legal rule combinations of the internal speeches is sequentially sequenced.
本 所提供的支持多 言統 和 音的方法及 統, 是在現有 休 各 的 上, 利用已有的軟件資源, 多 言播 的 和 。 尤其是在方言矣 較多的 , 可以 現在新增方言 不需更換 交換版本, 且 支持多方言 方言要求。 在此 上, 把相同的 法規則 未 行存儲 , 在 新的方言 只需要增 新方言的 基本 , 前預先存儲的 法規則 所述方言 行解析, 由此未 新的方言 有效的減少 工作量和提高功能 的 效率。 說明 The methods and systems supported by the Institute to support multi-language and audio are based on existing software resources and multi-speech. Especially in dialects, you can add new dialects without replacing the exchange version, and support multi-dialect dialect requirements. In this case, the same rule of law is not stored. In the new dialect, only the basics of the new dialect need to be added. The dialect line is parsed in the pre-stored rule, so that no new dialect effectively reduces the workload and improves. The efficiency of the function. Description
1是 統的 休 各 音流程。 1 is the system of the rest of the sound.
2是本 相 的 休 各 音方法流程 。 2 is the phase method of the phase sound method.
3力本 明媒休 各器多 言統 和 音的 統 。 休 方式 3 force Benming media to learn more than the system and the sound of the system. Hugh way
本 木方案的 內容是, 在 休 各 各中, 按照 言 分力 言 , 言 相 的 休 各 內部 言 , 通 控制 上 言 和 休 各 內部 言 的  The content of the project is that, in the various places, according to the words and deeds, the words of each of the internal words, the control of the words and the internal words of the rest
, 的不同控制 上 言 可以 到 休 各 內 部的 言或者多 言。 在此 上, 把 法規則 休 各 內部 言 相 , 可以 相同 法規則的 言 統 解析和組 合, 用戶只需指定 列表, 而 多 言的 和播 需求。  The different control of the words can go to the internal words or words. In this case, the internal rules of the law rule can be parsed and combined in the same legal rules. The user only needs to specify the list, and the content of the multicast and broadcast needs.
參照 2的本 的 休 各 音方法流程 , 方法主要 步驟201 休 各 將 按照 言 分力 言 , 所述 言 相 的 休 各 內部 言 , 所述 言 法規 則的  Refer to the method of the Hoop method of the second method, the main steps of the method 201, each of which will be in accordance with the words, the words of the internal words, the rules and regulations
例 休 各 在上海 音, 我們可以將普通 和上海 分 力 介 言 。 , 普通 和上海 就可以 相同的 休 各 內 言 和相同的 法規則。  For example, Xiu is in Shanghai, and we can introduce the general and Shanghai scores. , ordinary and Shanghai can have the same rest and the same rules of law.
步驟202 交換 統的控制 中 言 和所述 休 各 內 部 言 的 的  Step 202: Controlling the exchange system and the internal words of the rest
步驟203 交換 統下 控制 令播 指定的 , 所述控制 令中 控制 言 祇及 音信息  Step 203: Exchanging the control command specified by the command, and controlling the voice information only in the control command
步驟204 休 各 解析所述控制 令, 得 所述控制 令中的控 制 言 相 的 休 各 內部 言 , 所述 休 各 內部 言 的 言 得 的 法規則, 所述 法規則 所述 音信息 行解析 組合成新的 列表  Step 204: Each of the control commands is parsed, and the internal control words of the control statements in the control command are obtained, the legal rules of the spoken words of the internal words, and the audio information lines of the legal rules are analyzed and combined. New list
步驟205 解析出的 列表通 音 音。  Step 205 resolves the list of audio tones.
在本 的 介 中, 步驟 201 中所述將 按照 言 分力 言 , In this introduction, the description in step 201 will be followed. Divided into words,
將 介 相 近 的多 方言或者具有較少方言的 介 的 言分力 介 言 。  Introduce a multi-disciplinary or a dialect with a small number of dialects.
然, 言 的 可以 于多 考慮, 比 具有相同的 法規則 言 分力 介 , 或者具有地域相 性的 言 分力 介 等等, 休的神 分方法可以有多 考慮, 只要 本 的方法, 都可以 力本 。  However, the words can be considered more than the words with the same rules of law, or the words of the geographical phase, etc., Hugh’s method of dichotomy can be considered, as long as the method can this.
在本 的 介 中, 步驟 202 中所述 交換 統的控 制 中 言 和所述 休 各 內部 言 的 ,  In the present description, the control system of the exchange system described in step 202 and the internal words of the said
控制 下 的 言 介 言 相 的 休 各 內部 言 相 。  Under the control of the words and speeches of the various internal statements.
例 休 各 在上海 音, 我們可以將普通 和上海 介 言 , 然 將控制 令中的 言 ( Ch ) 休 各器 中 的 言, 就是普通 和上海 。 只要控制 令下 Ch 就 可以同 播 普通 和上海 了。  For example, Xiu is in Shanghai, we can introduce the ordinary and Shanghai, and then the words in the control order (Ch) will be the words of ordinary and Shanghai. As long as the control order is under Ch, you can broadcast both Normal and Shanghai.
在本 的 介 中, 所述方法近  In the context of this, the method is near
休 各 解析出控制 中的 言 多 休 各 內部 言 , 則 組合 列表, 多 言順序播 。 休 , 可 以在控制 令 音次數, 音次數 多 言順序播 。  Hugh analyzes the words in the control, and then combines the internal words, then combines the list and broadcasts in multiple languages. Hugh can be used to control the number of sounds and the number of sounds in a sequence.
其中所述步驟 202近 , 果 各 中需要 交換控制 的 言 解析成不同的方言播 , 典型的 中 各中播 提 示 般先播 普通 再播 其它方言, 中可以把 上 的 言 到 各內部的 言  In the step 202, the words that need to be exchanged are parsed into different dialects, and the typical ones in the typical broadcasts are first broadcasted and then re-broadcasted in other dialects, and the words can be put into the internal words.
其中所述步驟 204近 , 休 各 內部 言 預先 在 休 各 上的 的 列表, 得相 的 法規則 言 信息 行解析和組合。 于 多 內部 言 的 中, 只是解析 中 帶的 列表不同, 用的 法解析規則是同 介規則。 In the step 204, the list of the internal words in advance is in advance, and the rules of the law are obtained. Information line parsing and combination. In many internal statements, only the list in the parsing is different, and the parsing rules used are the same rules.
下面 本 的方法的  Following the method of this
步驟1, 在 休 各器中, 將 按照 言 , 休 各 內部的 言 。 般 介 相 近 的多 方言或者 介較少 方言的 介 的 言可以 。  Step 1, in the various devices, will be in accordance with the words, rest each internal words. A dialect with similar or similar dialects can be used.
步驟2,在 休 各器中配置 交換 統的控制 中下 的 言 和 休 各 內部 言 的 , 可以 所在地 的 方言在 休 各器中 配置。 般 控制 下 的 言  In step 2, in the control of the exchange system, the words in the control system and the words in the internal control are configured in the local dialect. General control
休 各器中 言。 Hugh in the various words.
步驟3, 交換 統下 控制 令播 指定的 , 令中包含 音的 言 , 音次數及 音內容。  In step 3, the exchange control system specifies the words, the number of sounds, and the sound content specified in the order.
步驟4, 休 各 收到所述控制 令, 控制 解析控制 令, 解析出 音的 言 , 音次數及 音內容。  In step 4, Hugh receives the control command, controls the analysis control command, and analyzes the speech, the number of times, and the content of the sound.
步驟5, 休 各 配置的控制 言 和 休 各 內部 言的 , 得需要 音的 各內部 言 。  Step 5, the control words of each configuration and the internal words of the rest, you need the internal words of the tone.
步驟6, 休 各 所述 言 相 的 法規則解析 組合 成新的 列表, 然 解析出的 列表和 音次數通 音 音。  In step 6, the legal rules of each of the phases are combined into a new list, and the list and the sound of the tone are parsed.
步驟7, 果步驟5解析出控制 中的 言 各內部多 言, 則在步驟6中 林組合 列表, 而 多 言的順序播 。  Step 7. If step 5 resolves the words in the control, each internal multi-word, in step 6, the forest combines the list, and the multi-sequence broadcasts in sequence.
其中步驟 2近 , 果在羊 的 家和 , 可以不必配置控制 的 言 各 內部 言 的 , 休 各 仍然按照 原先 好的 吋 音 果需要播 多 , 可以 需要 配置。  Step 2 is close, and if it is in the home of the sheep, it is not necessary to configure the internal words of the control. Hugh is still required to broadcast according to the original good sound, and can be configured.
參照 3 所示, 力本 明媒休 各器多 言統 和 音的 統 , 統 As shown in Figure 3, the power of the medium and the media System
矣 30, 于 休 各器中, 于將 按照 言矣 分力 言 矣, 言 矣相 的 休 各 內部 言 ,  矣 30, in the various devices of the rest, in accordance with the words and deeds, the words of the rest, the words of the rest,
言 矣 法規則的 交換 統的控制 中 言 和 休 各 內部 言 的  The exchange of the rules of the law and the control of the internal language
交換 統 31, 于下 控制 令播 指定的 , 控制 令中 控制 言 祇及 音信息 Switching system 31, in the control order specified by the control, in the control order, control words and audio information
32, 于 休 各 , 于解析 控制 令, 得 控制 令中的控制 言 相 的 休 各 內部 言 , 休 各 內部 言 的 言 矣 得 的 法規則, 法規則 音信息 行解析 組合成新的 列表。  32. Yu Xiu, in the analysis of the control order, the control words in the control order, the internal rules of the control words, the rules of the internal words, and the analysis of the rules.
在本 的 介 中, 32, 近 于根 解析出 的 列表通 音 音。  In this article, 32, near the root of the list of sounds.
在本 的 介 中, 矣 30, 步 于將 介 相 近 的多 方言或者具有較少方言的 介 的 言分力 介 言 矣。  In this book, 矣 30, step by step in the introduction of multiple dialects with similar or less dialects.
在本 的 介 中, 矣 30, 步 于 控制 下 的 言 介 言 矣相 的 休 各 內部 言 相 。  In this introduction, 矣 30, step by step under the control of the interpreter's internal words.
在本 的 介 中, 32, 步 于 休 則 組合 列表, 多 言順序播 。  In this article, 32, step in the combination of the rest of the list, in a multi-sequential broadcast.
下面 休 中的 本 明技木 行示 。  The following shows the skills of the technology in the rest.
以新增 方言矣 , 參考本 上 的步驟, 休 各 支持 新增方言的 步驟 下 步驟S , 新增 方言的基本 , 基本 法規則可以 用 前提 的 規則 In the new dialect, refer to the steps in this section, and each step under the support of adding dialects Step S, the basics of the new dialect, the rules of the basic law can use the rules of the premise
步驟S2, 將 方言按照 言 , 把 方言和最有可能 起 播 的方言 。 休 中, 最有可能 起播 的 言可以  Step S2, in accordance with the dialect, the dialect and the dialect most likely to be broadcast. In Hugh, the most likely words to broadcast
使用需要未  Need not to use
步驟S3, 在 休 各器中配置 , 步驟S2 , 將 方言 的 交換 統的控制 中下 的 言  Step S3, configured in the rest device, step S2, in the control of the exchange system of the dialect
步驟S4, 交換 統下 控制 令播 指定的 , 休同 新增方言的支持。  In step S4, the switching system controls the broadcaster to specify the support of the new dialect.
其中步驟S3中, 方言可以和 交換 統的控制 中下 的 言 。 也可以和多 交換 統的控制 中下 的 言 。 在播 只要播 任意 方言建立 的 言 , 就可以播 核方言。  In step S3, the dialect can be exchanged with the control of the exchange system. It can also be compared with the control of multiple exchange systems. In the broadcast, as long as the words established in any dialect are broadcast, the dialect can be broadcast.
其中步驟S4 休 , 休 各 收到 交換 統的 音信令, 把 播 的新增基本 按照所提 的 法規則 組合, 通 音 音。 的 , 通 配置未 各同的控制 的解析。 本 近 步 , 在配置上可以指定控制 言 和任何 或者多 方言 , 另外把相同的 法規則 , 組合中 方言 和 列表未 組合, 減少 工作量和 各功能 的重 。  In step S4, the rests receive the voice signaling of the switching system, and the new basics of the broadcast are combined according to the proposed rule, and the sound is sounded. The configuration of the control is not the same. In this step, you can specify the control word and any or more dialects in the configuration. In addition, the same rule, combined Chinese dialect and list are not combined, reducing the workload and the weight of each function.
以上所述 力本 的 而已, 不用于限制本 , 于本領 域的 木 未說, 本 可以有各 更 和 化。 凡在本 的精神和 原則 內, 所作的任何修 、 等同替換、 等, 包含在本 的 要求 固 內。  The above-mentioned methods are not used to limit the present, and the wood in this field has not been said to have been more versatile. Any revisions, equivalent substitutions, etc. made within the spirit and principles of this article are included in this requirement.

Claims

要求 Claim
1、 休 各器多 言統 和 音的方法, 其特 在于, 方 法  1. The method of multiple speech and sound, which is characterized by the method
休 各 按照 言 分力 言 , 所述 言 相 的 休 各 內部 言 祇及 所述 言 法規則的 交換 統的控制 中 言 和所述 休 各 內部 言 的  Hume is divided according to the words, and the internal words of the words are only in the control of the exchange of the rules of speech and the internal words of the various internal words.
交換 統下 控制 令播 , 所述控制 令中 控制 言 祇及 音信息  Exchanging the control command, the control command, the control message, and the sound information
休 各 解析所述控制 令, 得 所述控制 令中的控制 言 相 的 休 各 內部 言 , 所述 休 各 內部 言 的 言 得 的 法規則, 所述 法規則 所述 音信息 行解析 組合成新的 列表。  Each of the control commands is analyzed, and the internal control words of the control commands in the control command are obtained, and the legal rules of the internal words of the internal control words are analyzed, and the audio information lines of the legal rules are analyzed and combined into new ones. list of.
2、 要求1所述的方法, 其特 在于, 所述方法近 , 通 音 所述新的 列 音。  2. The method of claim 1 wherein the method is near and the new tone is spoken.
3、 要求1所述的方法, 其特 在于, 所述 休 各 按 照 言 分力 言 休  3. The method of claim 1, wherein the rest is in accordance with the statement
將 介 相 近 的多 方言或者具有較少方言的 介 的 言分力 介 言 。  Introduce a multi-disciplinary or a dialect with a small number of dialects.
4、 要求1所述的方法, 其特 在于, 所述 交換 統的控 制 中 言 和所述 休 各 內部 言 的 休  4. The method of claim 1, characterized in that the control system of the exchange system and the rest of the internal words of the rest
控制 下 的 言 介 言 相 的 休 各 內部 言 相 。 Under the control of the words and deeds of the various internal statements.
5、 要求1所述的方法, 其特 在于, 所述方法近 5. The method of claim 1 wherein the method is near
所述 休 各 解析出的所述控制 令中的控制 言  The control words in the control command parsed by the Hugh
多 休 各 內部 言 , 則 所述 休 各 內部 言 的 言 相 的 法規則組合 列表, 多 言順序播 。For each internal statement, the list of legal rules for the words of each internal statement is sequenced in multiple languages.
6、 休 各器多 言統 和 音的 統, 其特 在于, 裝 置  6, the various devices and sounds of the system, the special features, the device
, 于 休 各器中, 于將 按照 言 分力 言 , 所述 言 相 的 休 各 內部 言 祇及 所述 言 法規則的 交換 統的控制 中 言 和所述 休 各 內部 言 的  In the various devices, in accordance with the words of the words, the internal words of the words and the exchange of the rules of the rules of speech and the internal words of the various internal words
交換 統, 于下 控制 令播 , 所述控制 令中 控制 言 祇及 音信息  Switching the system, controlling the broadcast, and controlling the voice information in the control command
, 于 休 各 , 于解析所述控制 令, 得 所述 控制 令中的控制 言 相 的 休 各 內部 言 ,  , in Hugh, in the analysis of the control order, the control words in the control order, the internal words,
所述 休 各 內部 言 的 言 得 的 法規則, 所述 法規則 所述 音信息 行解析 組合成新的 列表。  The legal rules of the spoken words of the internal words, and the sound information rows of the legal rules are analyzed to form a new list.
7、 要求6所述的 統, 其特 在于,  7. The system described in claim 6 is characterized in that
8、 要求 6所述的 統, 其特 在于, 所述 將 言 分力 言 的方式 8. The system described in claim 6 is characterized by the manner in which the words are divided into words.
將 介 相 近 的多 方言或者具有較少方言的 介 的 言分力 介 言 。  Introduce a multi-disciplinary or a dialect with a small number of dialects.
9、 要求6所述的 統, 其特 在于, 所述 交換 統的控制 中 言 和所述 休 各 內部 言 的 的 方式 控制 下 的 言 介 言 相 的 休 各 內部 言 相 。 9. The system described in claim 6 is characterized in that the control system of the exchange system and the manner of the internal words of the rest Under the control of the words and deeds of the various internal statements.
10、 要求6 的 統, 其特 在于,  10. The system of requirement 6, which is characterized in that
若所述 解析出的所述控制 令中的控制 言 多 休 各 內部 言 , 則 所述 休 各 內部 言 的 言 相 的 法規則組合 列表, 多 言順序播 。  If the control in the parsed control command has multiple internal statements, the list of legal rule combinations of the internal words is sequentially sequenced.
PCT/CN2010/073282 2009-12-15 2010-05-26 Method and system for unified multi-language development and voice play in media server WO2010148893A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN 200910259266 CN101729570A (en) 2009-12-15 2009-12-15 Multilingual unified development and playback method and system of media server
CN200910259266.7 2009-12-15

Publications (1)

Publication Number Publication Date
WO2010148893A1 true WO2010148893A1 (en) 2010-12-29

Family

ID=42449768

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2010/073282 WO2010148893A1 (en) 2009-12-15 2010-05-26 Method and system for unified multi-language development and voice play in media server

Country Status (2)

Country Link
CN (1) CN101729570A (en)
WO (1) WO2010148893A1 (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102014214428A1 (en) * 2014-07-23 2016-01-28 Bayerische Motoren Werke Aktiengesellschaft Improvement of speech recognition in a vehicle
KR102384641B1 (en) * 2017-02-20 2022-04-08 엘지전자 주식회사 Method for controlling an intelligent system that performs multilingual processing
CN109817208A (en) * 2019-01-15 2019-05-28 上海交通大学 A kind of the driver's speech-sound intelligent interactive device and method of suitable various regions dialect

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5940790A (en) * 1994-02-14 1999-08-17 Nokia Telecommunications Oy Multilingual operation and maintenance interface for a telecommunication exchange
CN1377193A (en) * 2001-03-28 2002-10-30 华为技术有限公司 Multi-speech prompting method for intelligent network
CN101001286A (en) * 2006-12-27 2007-07-18 中兴通讯股份有限公司 Method and system for customing prompt voice speech in digital program control exchange
CN101137077A (en) * 2007-10-16 2008-03-05 中兴通讯股份有限公司 Digital program controlled exchanger and method to support multilingual audio playback

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5940790A (en) * 1994-02-14 1999-08-17 Nokia Telecommunications Oy Multilingual operation and maintenance interface for a telecommunication exchange
CN1377193A (en) * 2001-03-28 2002-10-30 华为技术有限公司 Multi-speech prompting method for intelligent network
CN101001286A (en) * 2006-12-27 2007-07-18 中兴通讯股份有限公司 Method and system for customing prompt voice speech in digital program control exchange
CN101137077A (en) * 2007-10-16 2008-03-05 中兴通讯股份有限公司 Digital program controlled exchanger and method to support multilingual audio playback

Also Published As

Publication number Publication date
CN101729570A (en) 2010-06-09

Similar Documents

Publication Publication Date Title
Liu et al. Hkust/mts: A very large scale mandarin telephone speech corpus
US8082142B2 (en) Speech recognition method, speech recognition system and server thereof
TWI249729B (en) Voice browser dialog enabler for a communication system
US7450698B2 (en) System and method of utilizing a hybrid semantic model for speech recognition
CN103377028B (en) For with the method and system of vice activation man-machine interface
US20140358516A1 (en) Real-time, bi-directional translation
US20060161431A1 (en) System and method for independently recognizing and selecting actions and objects in a speech recognition system
Takezawa et al. Multilingual spoken language corpus development for communication research
CN101847406B (en) Speech recognition query method and system
DE60143797D1 (en) VOICE RECOGNITION
JP2014071769A (en) Expression conversion device, method and program
US20100017193A1 (en) Method, spoken dialog system, and telecommunications terminal device for multilingual speech output
JP2017120616A (en) Machine translation method and machine translation system
WO2010148893A1 (en) Method and system for unified multi-language development and voice play in media server
JP2004518195A (en) Automatic dialogue system based on database language model
KR20110080096A (en) Dialog system using extended domain and natural language recognition method thereof
Durston et al. OASIS natural language call steering trial
US20060265225A1 (en) Method and apparatus for voice recognition
US6501751B1 (en) Voice communication with simulated speech data
US6662157B1 (en) Speech recognition system for database access through the use of data domain overloading of grammars
US11790913B2 (en) Information providing method, apparatus, and storage medium, that transmit related information to a remote terminal based on identification information received from the remote terminal
AT&T
CN102800313A (en) Method for supporting multi-voice recognition engine in Voice extensible markup language (XML) 2.0
Marasek et al. Design and Data Collection for Spoken Polish Dialogs Database.
JP7422702B2 (en) Barrier-free smart voice system and its control method

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 10791399

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 5213/DELNP/2012

Country of ref document: IN

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 10791399

Country of ref document: EP

Kind code of ref document: A1