CN104732969A - Voice processing system and method - Google Patents
Voice processing system and method Download PDFInfo
- Publication number
- CN104732969A CN104732969A CN201310714903.1A CN201310714903A CN104732969A CN 104732969 A CN104732969 A CN 104732969A CN 201310714903 A CN201310714903 A CN 201310714903A CN 104732969 A CN104732969 A CN 104732969A
- Authority
- CN
- China
- Prior art keywords
- spokesman
- speech
- voice
- personal information
- section
- 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.)
- Pending
Links
Landscapes
- Reverberation, Karaoke And Other Acoustics (AREA)
Abstract
Disclosed is a voice processing method. The method comprises the steps of extracting voice features corresponding to a speaker from voices generated by the speaker within a continuous period of time, sequentially extracting voice features of all speakers from voices generated by the speakers within the continuous period of time according to the speaking time sequence of the speakers; sequentially judging whether the extracted voice features are matched with a voice pattern model in a voice pattern database or not; if yes, sequentially converting the voices generated by each speaker within the continuous period of time into a passage; inserting personal information corresponding to the voice pattern model in front of the passage converted from the voices of the corresponding speaker, generating a text containing the passage and the personal information, and sequentially adding the corresponding personal information and the passages converted from the voices generated by the speakers within the continuous period of time into the corresponding texts according to the speaking time sequence of the speakers. The invention further provides a voice processing system. The generated passages are easy to understand.
Description
Technical field
The present invention relates to speech processing system and method for speech processing.
Background technology
At present, the voice that meeting or interview are recorded can be automatically converted to word.But owing to there is no the personal information of corresponding spokesman before generated word, thus make the word of this generation hard to understand, the word arrangement that inconvenient keeper is follow-up.
Summary of the invention
In view of above content, be necessary to provide a kind of speech processing system and method for speech processing, the word of this generation can be made understandable.
A kind of speech processing system, this speech processing system comprises: a feature acquisition module, for extracting phonetic feature corresponding to spokesman in the voice of the spokesman in a sound collection unit collection in a continuous time section, and according to the time order and function of each spokesman speech, extract the phonetic feature of each spokesman in the voice in continuous time section successively, one sound identification module, for judging whether the phonetic feature of this extraction mates with the sound-groove model in a voiceprint data storehouse, and wherein, each sound-groove model in this voiceprint data storehouse is associated with the personal information of a correspondence successively, this spokesman, when the phonetic feature for extraction before determining to deserve mates with the sound-groove model in voiceprint data storehouse, is passage in the speech conversion in this section by voice conversion module successively continuous time, and a text generation module, for inserting personal information corresponding to this sound-groove model before this section of word of the speech conversion this spokesman, and generate the text that comprises this section of word and personal information, and after inserting personal information corresponding to this sound-groove model successively before the passage of the speech conversion of next spokesman, the passage change next spokesman's voice and the personal information of correspondence are increased in the text successively, and according to the time order and function that each spokesman makes a speech, successively the personal information of passage that the continuous time of each spokesman, the voice in section were changed and correspondence is increased in the text.
A kind of method of speech processing, the method comprises: in voice in a continuous time section of a spokesman of a sound collection unit collection, extract phonetic feature corresponding to spokesman, and according to the time order and function of each spokesman speech, extract the phonetic feature of each spokesman in the voice in continuous time section successively, judge whether the phonetic feature of this extraction mates with the sound-groove model in a voiceprint data storehouse, and wherein, each sound-groove model in this voiceprint data storehouse is associated with the personal information of a correspondence successively, when before determining to deserve, the phonetic feature of extraction mates with the sound-groove model in voiceprint data storehouse, be passage by this spokesman in the speech conversion in this section successively continuous time, and before this section of word of the speech conversion of this spokesman, insert personal information corresponding to this sound-groove model, and generate the text that comprises this section of word and personal information, and after inserting personal information corresponding to this sound-groove model successively before the passage of the speech conversion of next spokesman, the passage change next spokesman's voice and corresponding personal information are increased in the text successively, and according to the time order and function that each spokesman makes a speech, successively the personal information of passage that the continuous time of each spokesman, the voice in section were changed and correspondence is increased in the text.
The present invention is by being first passage by the speech conversion of the continuous time of a spokesman in section, before personal information being inserted in this section of word, generate the text that comprises this section of word and personal information, and according to the time order and function that each spokesman makes a speech, the passage changed by the voice of each spokesman successively and personal information are increased in the text, thus make the word of this generation understandable.
Accompanying drawing explanation
Fig. 1 is the block diagram of speech processing system in an embodiment of the present invention.
Fig. 2 is the process flow diagram of method of speech processing in an embodiment of the present invention.
Main element symbol description
Speech processing system | 10 |
Voice processing apparatus | 1 |
Input block | 2 |
Sound collection unit | 3 |
Central processing unit | 20 |
Storer | 30 |
Feature acquisition module | 11 |
Sound identification module | 12 |
Voice conversion module | 13 |
Text generation module | 14 |
Following embodiment will further illustrate the present invention in conjunction with above-mentioned accompanying drawing.
Embodiment
Referring to Fig. 1, is the block diagram of the speech processing system 10 of an embodiment of the present invention.In the present embodiment, this speech processing system 10 is installed and is run in a voice processing apparatus 1, for the speech conversion of each spokesman is become word, and before personal information corresponding for each spokesman being inserted in corresponding word, thus identify spokesman's identity that word that this speech conversion becomes is corresponding.Described voice processing apparatus 1 is connected with input block 2 and a sound collection unit 3, and this voice processing apparatus 1 also comprises a central processing unit (Central Processing Unit, CPU) 20 and a storer 30.
In the present embodiment, this speech processing system 10 comprises feature acquisition module 11, sound identification module 12, voice conversion module 13 and a text generation module 14.Module alleged by the present invention refer to a kind of can be performed and the series of computation machine program block of specific function can be completed by the central processing unit 20 of voice processing apparatus 1, it is stored in the storer 30 of voice processing apparatus 1.Wherein, in this storer 30, also store voiceprint data storehouse and voice document, store the personal information of spokesman corresponding to the sound-groove model of spokesman and this sound-groove model in this voiceprint data storehouse, as name, photo etc.This voice document is the audio file comprising the speech record of each spokesman of record.
This feature acquisition module 11 is for extracting phonetic feature corresponding to spokesman in voice in a continuous time section of the spokesman that gathers in this sound collection unit 3, and according to the time order and function of each spokesman speech, extract the phonetic feature of each spokesman in the voice in continuous time section successively.Wherein, in one embodiment, this feature acquisition module 11, when sound collection unit 3 gathers spokesman, synchronously carries out the extraction of spokesman's phonetic feature.In another embodiment, this feature acquisition module 11 responds user selects voice document operation by this input block 2, extracts the phonetic feature of each spokesman from the voice document of this selection successively.In the present embodiment, this feature acquisition module 11 carries out the extraction of the phonetic feature of spokesman by Mel Cepstral Frequency Coefficients.But the present invention extracts phonetic feature and is not limited to aforesaid way, other extract phonetic feature and are also included within the scope disclosed by the present invention.
This sound identification module 12 is for judging whether the phonetic feature of the spokesman that this feature acquisition module 11 extracts mates with the sound-groove model in this voiceprint data storehouse successively.
Spokesman, for when determining that this phonetic feature mates with the sound-groove model in this voiceprint data storehouse, is passage in the speech conversion in this section by this voice conversion module 13 successively continuous time.
Text generation module 14 is for inserting personal information corresponding to this sound-groove model before this section of word of the speech conversion spokesman, generate the text that comprises this section of word and personal information, and after inserting personal information corresponding to this sound-groove model successively before the passage of the speech conversion of next spokesman, the passage change next spokesman's voice and corresponding personal information are increased in the text successively, and according to the time order and function that each spokesman makes a speech, successively the personal information of passage that the continuous time of each spokesman, the voice in section were changed and correspondence is increased in the text.
Such as, when this sound collection unit 3 gathers, Li Li makes a speech at 9:00-9:05, and Lei Lin makes a speech at 9:05-9:10, and Zhang Mei makes a speech at 9:05-9:08, then beautifully makes a speech at 9:10-9:15.This feature acquisition module first obtains beautiful phonetic feature.This voice conversion module 13 when this sound identification module 12 determines that the phonetic feature of this extraction mates with the sound-groove model in voiceprint data storehouse, is converted to passage in 9:00-9:05 speech by beautiful " I is hungry by belly, you have and eat? "Text generation module 14 generates text " Li Li: I is hungry by belly, you have and eat? "Then, this feature acquisition module 11, this sound identification module 12, this voice conversion module 13 perform corresponding operation successively, and the personal information of thunder woods, Zhang Mei, section word that beautiful voice are changed and correspondence is then increased in the text by text generation module successively.
In the present embodiment, this feature acquisition module 11 is for judging whether the phonetic feature obtained before and after this changes.When the phonetic feature obtained before and after this changes, this feature acquisition module 11 determines that the phonetic feature of this acquisition is the speech of different spokesman, thus determine that the speech of a spokesman terminates, and determine the voice that the voice gathered in the time period of this sound collection unit 3 before this terminates are this spokesman in continuous time section.When the phonetic feature obtained before and after this feature acquisition module 11 judges does not change, this feature acquisition module 11 determines that the phonetic feature obtained before and after this is the speech of same spokesman, namely, determine to be still the speech of this spokesman in continuous time section, and determine the phonetic feature of this spokesman.
In the present embodiment, text generation module 14 is when being increased to the section word of each spokesman and personal information in the text successively, the passage first changed by next spokesman's voice and the personal information of correspondence restart new paragraph, be increased to again in the text, and according to the time order and function that each spokesman makes a speech, the passage changed by each spokesman's voice successively and the personal information of correspondence restart new paragraph and are increased in the text.As the content recorded in the first paragraph for " Li Li: I is hungry by belly; you have and eat? " the content recorded in second segment for " Lei Lin: I has chocolate; you will eat? " the content recorded in the 3rd section how be " Zhang Mei: milk;? ", the content recorded in the 4th section be " Li Li: genuine? chocolate? excellent, that is my favourite thing just.”。Then, keeper can determine the speech of each spokesman easily.
In the present embodiment, text generation module 14 also for recording the speech initial time of each spokesman in each section continuous time, and before the initial time of this record being inserted respectively each section of corresponding word.Thus this keeper can know the speech sequencing of each spokesman and concrete time limit of speech.
Please refer to Fig. 2, is the process flow diagram of the method for speech processing of an embodiment of the present invention.
In step s 201, phonetic feature corresponding to spokesman is extracted in the voice of the spokesman that this feature acquisition module 11 gathers in this sound collection unit 3 in a continuous time section, and according to the time order and function of each spokesman speech, extract the phonetic feature of each spokesman in the voice in continuous time section successively.
In step S202, this sound identification module 12 judges whether the phonetic feature of the spokesman that this feature acquisition module 11 extracts mates with the sound-groove model in this voiceprint data storehouse successively.When the phonetic feature of the spokesman that this feature acquisition module 11 extracts mates with the sound-groove model in this voiceprint data storehouse, perform step S203.When the phonetic feature of the spokesman that this feature acquisition module 11 extracts does not mate with any sound-groove model in this voiceprint data storehouse, perform step S201.
In step S203, spokesman is passage in the speech conversion in continuous time section by this voice conversion module 13 successively.
In step S204, text generation module 14 inserts personal information corresponding to this sound-groove model before this section of word of the speech conversion of spokesman, generate the text that comprises this section of word and personal information, and after inserting personal information corresponding to this sound-groove model successively before the passage of the speech conversion of next spokesman, the passage change next spokesman's voice and the personal information of correspondence are increased in the text successively, and according to the time order and function that each spokesman makes a speech, successively the personal information of passage that the continuous time of each spokesman, the voice in section were changed and correspondence is increased in the text.
To those skilled in the art, the actual needs can produced according to scheme of the invention of the present invention and inventive concept combination is made other and is changed accordingly or adjustment, and these changes and adjustment all should belong to the protection domain of the claims in the present invention.
Claims (8)
1. a speech processing system, is characterized in that, this speech processing system comprises:
One feature acquisition module, for extracting phonetic feature corresponding to spokesman in the voice of the spokesman in a sound collection unit collection in a continuous time section, and according to the time order and function of each spokesman speech, extract the phonetic feature of each spokesman in the voice in continuous time section successively;
One sound identification module, for judging whether the phonetic feature of this extraction mates with the sound-groove model in a voiceprint data storehouse, and wherein, each sound-groove model in this voiceprint data storehouse is associated with the personal information of a correspondence successively;
This spokesman, when the phonetic feature for extraction before determining to deserve mates with the sound-groove model in voiceprint data storehouse, is passage in the speech conversion in this section by voice conversion module successively continuous time; And
One text generation module, for inserting personal information corresponding to this sound-groove model before this section of word of the speech conversion this spokesman, and generate the text that comprises this section of word and personal information, and after inserting personal information corresponding to this sound-groove model successively before the passage of the speech conversion of next spokesman, the passage change next spokesman's voice and the personal information of correspondence are increased in the text successively, and according to the time order and function that each spokesman makes a speech, successively the personal information of passage that the continuous time of each spokesman, the voice in section were changed and correspondence is increased in the text.
2. speech processing system as claimed in claim 1, is characterized in that: this feature acquisition module is for judging whether the phonetic feature obtained before and after this changes; When feature acquisition module judges that the phonetic feature obtained before and after this changes, confirm that the speech of a spokesman terminates, and determine the voice that the voice gathered in the time period of this sound collection unit before this terminates are this spokesman in continuous time section; This feature acquisition module, when judging that the phonetic feature that obtains when this before and after does not change, determines the speech being still same spokesman in continuous time section.
3. speech processing system as claimed in claim 1, it is characterized in that: text generation module is when being increased to the section word of each spokesman and personal information in the text successively, the passage first changed by next spokesman's voice and the personal information of correspondence restart new paragraph, be increased to again in the text, and according to the time order and function that each spokesman makes a speech, the passage changed by each spokesman's voice successively and the personal information of correspondence restart new paragraph and are increased in the text.
4. speech processing system as claimed in claim 1, is characterized in that: text generation module also for recording the speech initial time of each spokesman in each section continuous time, and before the initial time of this record being inserted respectively each section of corresponding word.
5. a method of speech processing, is characterized in that, the method comprises:
In voice in a continuous time section of a spokesman of a sound collection unit collection, extract phonetic feature corresponding to spokesman, and according to the time order and function of each spokesman's speech, extract the phonetic feature of each spokesman in the voice in continuous time section successively;
Judge whether the phonetic feature of this extraction mates with the sound-groove model in a voiceprint data storehouse, and wherein, each sound-groove model in this voiceprint data storehouse is associated with the personal information of a correspondence successively;
When before determining to deserve, the phonetic feature of extraction mates with the sound-groove model in voiceprint data storehouse, be passage by this spokesman in the speech conversion in this section successively continuous time; And
Personal information corresponding to this sound-groove model is inserted before this section of word of the speech conversion of this spokesman, and generate the text that comprises this section of word and personal information, and after inserting personal information corresponding to this sound-groove model successively before the passage of the speech conversion of next spokesman, the passage change next spokesman's voice and corresponding personal information are increased in the text successively, and according to the time order and function of each spokesman speech, successively the personal information of passage that the continuous time of each spokesman, the voice in section were changed and correspondence is increased in the text.
6. method of speech processing as claimed in claim 5, it is characterized in that, the method comprises:
Judge whether the phonetic feature obtained before and after this changes;
When the phonetic feature obtained before and after this changes, confirm that the speech of a spokesman terminates, and determine the voice that the voice gathered in the time period of this sound collection unit before this terminates are this spokesman in continuous time section; And
When the phonetic feature obtained before and after this does not change, determine still to be the speech of same spokesman in continuous time section.
7. method of speech processing as claimed in claim 5, it is characterized in that, the method comprises:
When successively the section word of each spokesman and personal information being increased in the text, the passage first changed by next spokesman's voice and the personal information of correspondence restart new paragraph, be increased to again in the text, and according to the time order and function that each spokesman makes a speech, the passage changed by each spokesman's voice successively and the personal information of correspondence restart new paragraph and are increased in the text.
8. method of speech processing as claimed in claim 5, it is characterized in that, the method comprises:
Record the speech initial time of each spokesman in each section continuous time, and before the initial time of this record being inserted respectively each section of corresponding word.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310714903.1A CN104732969A (en) | 2013-12-23 | 2013-12-23 | Voice processing system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310714903.1A CN104732969A (en) | 2013-12-23 | 2013-12-23 | Voice processing system and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104732969A true CN104732969A (en) | 2015-06-24 |
Family
ID=53456809
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310714903.1A Pending CN104732969A (en) | 2013-12-23 | 2013-12-23 | Voice processing system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104732969A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105427857A (en) * | 2015-10-30 | 2016-03-23 | 华勤通讯技术有限公司 | Method and system used for generating text records |
CN105488227A (en) * | 2015-12-29 | 2016-04-13 | 惠州Tcl移动通信有限公司 | Electronic device and method for processing audio file based on voiceprint features through same |
CN106453865A (en) * | 2016-09-27 | 2017-02-22 | 努比亚技术有限公司 | Mobile terminal and voice-text converting method |
CN106653002A (en) * | 2016-12-15 | 2017-05-10 | 咪咕数字传媒有限公司 | Literal live broadcasting method and platform |
CN107437415A (en) * | 2017-08-09 | 2017-12-05 | 科大讯飞股份有限公司 | A kind of intelligent sound exchange method and system |
CN108053828A (en) * | 2017-12-25 | 2018-05-18 | 无锡小天鹅股份有限公司 | Determine the method, apparatus and household electrical appliance of control instruction |
CN109147796A (en) * | 2018-09-06 | 2019-01-04 | 平安科技(深圳)有限公司 | Audio recognition method, device, computer equipment and computer readable storage medium |
CN109309751A (en) * | 2017-07-28 | 2019-02-05 | 腾讯科技(深圳)有限公司 | Voice recording method, electronic equipment and storage medium |
CN109344232A (en) * | 2018-11-13 | 2019-02-15 | 平安科技(深圳)有限公司 | A kind of public feelings information search method and terminal device |
CN109887508A (en) * | 2019-01-25 | 2019-06-14 | 广州富港万嘉智能科技有限公司 | A kind of meeting automatic record method, electronic equipment and storage medium based on vocal print |
CN110677614A (en) * | 2019-10-15 | 2020-01-10 | 广州国音智能科技有限公司 | Information processing method, device and computer readable storage medium |
CN111770300A (en) * | 2020-06-24 | 2020-10-13 | 北京安博创赢教育科技有限责任公司 | Conference information processing method and virtual reality head-mounted equipment |
-
2013
- 2013-12-23 CN CN201310714903.1A patent/CN104732969A/en active Pending
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105427857B (en) * | 2015-10-30 | 2019-11-08 | 华勤通讯技术有限公司 | Generate the method and system of writing record |
CN105427857A (en) * | 2015-10-30 | 2016-03-23 | 华勤通讯技术有限公司 | Method and system used for generating text records |
CN105488227A (en) * | 2015-12-29 | 2016-04-13 | 惠州Tcl移动通信有限公司 | Electronic device and method for processing audio file based on voiceprint features through same |
CN106453865A (en) * | 2016-09-27 | 2017-02-22 | 努比亚技术有限公司 | Mobile terminal and voice-text converting method |
CN106653002A (en) * | 2016-12-15 | 2017-05-10 | 咪咕数字传媒有限公司 | Literal live broadcasting method and platform |
CN109309751A (en) * | 2017-07-28 | 2019-02-05 | 腾讯科技(深圳)有限公司 | Voice recording method, electronic equipment and storage medium |
CN109309751B (en) * | 2017-07-28 | 2021-08-06 | 腾讯科技(深圳)有限公司 | Voice recording method, electronic device and storage medium |
WO2019029352A1 (en) * | 2017-08-09 | 2019-02-14 | 科大讯飞股份有限公司 | Intelligent voice interaction method and system |
CN107437415B (en) * | 2017-08-09 | 2020-06-02 | 科大讯飞股份有限公司 | Intelligent voice interaction method and system |
CN107437415A (en) * | 2017-08-09 | 2017-12-05 | 科大讯飞股份有限公司 | A kind of intelligent sound exchange method and system |
CN108053828A (en) * | 2017-12-25 | 2018-05-18 | 无锡小天鹅股份有限公司 | Determine the method, apparatus and household electrical appliance of control instruction |
CN109147796A (en) * | 2018-09-06 | 2019-01-04 | 平安科技(深圳)有限公司 | Audio recognition method, device, computer equipment and computer readable storage medium |
CN109147796B (en) * | 2018-09-06 | 2024-02-09 | 平安科技(深圳)有限公司 | Speech recognition method, device, computer equipment and computer readable storage medium |
CN109344232A (en) * | 2018-11-13 | 2019-02-15 | 平安科技(深圳)有限公司 | A kind of public feelings information search method and terminal device |
CN109344232B (en) * | 2018-11-13 | 2024-03-15 | 平安科技(深圳)有限公司 | Public opinion information retrieval method and terminal equipment |
CN109887508A (en) * | 2019-01-25 | 2019-06-14 | 广州富港万嘉智能科技有限公司 | A kind of meeting automatic record method, electronic equipment and storage medium based on vocal print |
CN110677614A (en) * | 2019-10-15 | 2020-01-10 | 广州国音智能科技有限公司 | Information processing method, device and computer readable storage medium |
CN111770300A (en) * | 2020-06-24 | 2020-10-13 | 北京安博创赢教育科技有限责任公司 | Conference information processing method and virtual reality head-mounted equipment |
CN111770300B (en) * | 2020-06-24 | 2022-07-05 | Oook(北京)教育科技有限责任公司 | Conference information processing method and virtual reality head-mounted equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104732969A (en) | Voice processing system and method | |
US10593332B2 (en) | Diarization using textual and audio speaker labeling | |
CN109065031B (en) | Voice labeling method, device and equipment | |
US10977299B2 (en) | Systems and methods for consolidating recorded content | |
CN108305643B (en) | Method and device for determining emotion information | |
JP6556575B2 (en) | Audio processing apparatus, audio processing method, and audio processing program | |
US20160189713A1 (en) | Apparatus and method for automatically creating and recording minutes of meeting | |
CN105096941A (en) | Voice recognition method and device | |
WO2009101837A1 (en) | Mark insertion device and mark insertion method | |
CN111128223A (en) | Text information-based auxiliary speaker separation method and related device | |
CN104078044A (en) | Mobile terminal and sound recording search method and device of mobile terminal | |
US20160189107A1 (en) | Apparatus and method for automatically creating and recording minutes of meeting | |
US20160189103A1 (en) | Apparatus and method for automatically creating and recording minutes of meeting | |
CN110675866B (en) | Method, apparatus and computer readable recording medium for improving at least one semantic unit set | |
CN104750677A (en) | Speech translation apparatus, speech translation method and speech translation program | |
CN104639742A (en) | Method and device for assisting in learning speaking through mobile terminal | |
CN113782026A (en) | Information processing method, device, medium and equipment | |
CN106710587A (en) | Speech recognition data pre-processing method | |
CN108899016B (en) | Voice text normalization method, device and equipment and readable storage medium | |
JP6322125B2 (en) | Speech recognition apparatus, speech recognition method, and speech recognition program | |
CN115831125A (en) | Speech recognition method, device, equipment, storage medium and product | |
Yarra et al. | Automatic intonation classification using temporal patterns in utterance-level pitch contour and perceptually motivated pitch transformation | |
CN112837688A (en) | Voice transcription method, device, related system and equipment | |
TW201530535A (en) | Speech processing system and speech processing method | |
JP6078402B2 (en) | Speech recognition performance estimation apparatus, method and program thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150624 |