CN108597504A - A kind of speech analysis method, apparatus and robot - Google Patents
A kind of speech analysis method, apparatus and robot Download PDFInfo
- Publication number
- CN108597504A CN108597504A CN201810300973.5A CN201810300973A CN108597504A CN 108597504 A CN108597504 A CN 108597504A CN 201810300973 A CN201810300973 A CN 201810300973A CN 108597504 A CN108597504 A CN 108597504A
- Authority
- CN
- China
- Prior art keywords
- time
- date
- related information
- unit
- value
- 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
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/18—Speech classification or search using natural language modelling
- G10L15/1822—Parsing for meaning understanding
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
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)
- Signal Processing (AREA)
- Artificial Intelligence (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of speech analysis method, apparatus and robots, obtain phonetic order input by user, comprising after time related information in determining the phonetic order, date value and time numerical value are parsed from the time related information, by merging date value and time numerical value, a precise time is obtained.Therefore, for comparing traditional scheme, the invention enables speech analysis schemes to be no longer limited to date scene domain with scene, but can also be used in Novel Temporal Scenario range, so that the utilization scene of speech analysis scheme is no longer single.
Description
Technical field
The present invention relates to speech analysis technical fields, more specifically, being related to a kind of speech analysis method, apparatus and machine
People.
Background technology
With the leap at full speed of science and technology, robot is widely used in multiple fields, and gradually develops to intelligent direction.
Currently, robot can carry out voice messaging parsing according to phonetic order input by user.For example, pre- in train ticket, plane ticket
It orders etc. under scenes, would generally include the keywords such as ticket booking date in phonetic order input by user, such as today, tomorrow, Monday, the moon
The keyword of the non-standard time format such as bottom, for another example key of the standard time formats such as January 1, in January, 2018 in 2018
Word, robot is by parsing phonetic order input by user, you can obtains the ticket booking date of user.
However, traditional robot when parsing phonetic order input by user, be merely able to be resolved to such as today,
Tomorrow, the day after tomorrow, what day etc. noncanonical formats, or include the date range of the reference formats such as year, month, day, analytical range is not
It is enough wide, cause it to be limited only in date scene domain with scene.
Invention content
In view of this, the present invention discloses a kind of speech analysis method, apparatus and robot, to realize speech analysis scheme
It is no longer limited to date scene domain with scene, but Novel Temporal Scenario range can also be operated in.
A kind of speech analysis method, including：
Obtain phonetic order input by user；
Judge in the phonetic order whether to include time related information；
If it is, parsing date value from the time related information；
Time numerical value is parsed from the time related information；
The date value and the time numerical value are merged, precise time is obtained.
Preferably, the process that date value is parsed from the time related information, specifically includes：
Judge whether the time related information is the interval date；
If the time related information includes the interval date, the interval date is determined as the day issue
Value；
If the time related information do not include the interval date, judge the time related information whether include
Weekly data；
If the time related information includes the weekly data, the weekly data is determined as the date value；
If the time related information does not include the weekly data, judge whether the time related information includes to work as
The difference of the number of days of the nearest date on preceding date, the nearest date and the current date is less than threshold value；
If the time related information includes the nearest date, the nearest date is determined as the day issue
Value；
If the time related information do not include the nearest date, judge the time related information whether include
The reference format date；
If the time correlation date includes the reference format date, the reference format date is determined as institute
State date value；
If the time correlation date does not include the reference format date, judge that the time related information does not wrap
Containing the date value.
Preferably, the process that time numerical value is parsed from the time related information, specifically includes：
Judge whether the time related information includes interval time；
If the time related information includes the interval time, the interval time is determined as the time number
Value；
If the time related information do not include the interval time, judge the time related information whether include
The reference format time；
If the time related information includes the reference format time, the reference format time is determined as institute
State time numerical value；
If the time related information does not include the reference format time, whether the time related information is judged
Including summarize the time, the summarys time be characterization a period of time, and not comprising when, divide and the second time；
If the time related information includes the summary time, the summary time is determined as the time number
Value；
If the time related information does not include the summary time, judge not including in the time related information
The time numerical value.
Preferably, it is merged by the date value and the time numerical value, obtains precise time, specifically include：
When the time numerical value is earlier than current time, then the date value is adjusted to the next of the date value
Date value；
The time numerical value and next date value are merged, the precise time is obtained.
Preferably, further include：
Export the precise time.
A kind of speech analysis device, including：
Acquiring unit, for obtaining phonetic order input by user；
Judging unit, for judging in the phonetic order whether to include time related information；
First resolution unit, in the case where the judging unit is judged as YES, from the time related information
Parse date value；
Second resolution unit, for parsing time numerical value from the time related information；
Combining unit obtains precise time for merging the date value and the time numerical value.
Preferably, first resolution unit specifically includes：
First judgment sub-unit, for judging whether the time related information is the interval date；
First determination subelement, in the case where first judgment sub-unit is judged as YES, will it is described between the next day
Phase is determined as the date value；
Second judgment sub-unit, in the case where first judgment sub-unit is judged as NO, judging the time
Whether relevant information includes weekly data；
Second determination subelement, in the case where second judgment sub-unit is judged as YES, by the weekly data
It is determined as the date value；
Third judgment sub-unit, in the case where second judgment sub-unit is judged as NO, judging the time
Whether relevant information includes the nearest date of current date, and the difference of the number of days of the nearest date and the current date is less than threshold
Value；
Third determination subelement, in the case where the third judgment sub-unit is judged as YES, will described in most recently
Phase is determined as the date value；
4th judgment sub-unit, in the case where the third judgment sub-unit is judged as NO, judging the time
Whether relevant information includes the reference format date；
4th determination subelement, in the case where the 4th judgment sub-unit is judged as YES, by the reticle
The formula date is determined as the date value；
First judging unit, in the case where the 4th judgment sub-unit is judged as NO, judging the time phase
It closes information and does not include the date value.
Preferably, second resolution unit specifically includes：
5th judgment sub-unit, for judging whether the time related information includes interval time；
5th determination subelement, in the case where the 5th judgment sub-unit is judged as YES, when by the interval
Between be determined as the time numerical value；
6th judgment sub-unit, in the case where the 5th judgment sub-unit is judged as NO, judging the time
Whether relevant information includes the reference format time；
6th determination subelement, in the case where the 6th judgment sub-unit is judged as YES, by the reticle
The formula time is determined as the time numerical value；
7th judgment sub-unit, in the case where the 6th judgment sub-unit is judged as NO, judging the time
Relevant information whether comprising summarize the time, the summarys time be characterization a period of time, and not comprising when, divide and the second time；
7th determination subelement, in the case where the 7th judgment sub-unit is judged as YES, when by the summary
Between be determined as the time numerical value；
Second judgement subelement, in the case where the 7th judgment sub-unit is judged as NO, judging the time
The time numerical value is not included in relevant information.
Preferably, the combining unit specifically includes：
Subelement is adjusted, for working as the time numerical value earlier than current time, then the date value is adjusted to described
Next date value of date value；
Merge subelement and obtains the essence for merging the time numerical value and next date value
The true time.
Preferably, further include：
Output unit, for exporting the precise time.
A kind of robot, including speech analysis device described above.
From above-mentioned technical solution it is found that the invention discloses a kind of speech analysis method, apparatus and robot, obtains and use
The phonetic order of family input is parsed comprising after time related information in determining the phonetic order from the time related information
Go out date value and time numerical value, by merging date value and time numerical value, obtains a precise time.Therefore,
For traditional scheme, the invention enables speech analysis schemes to be no longer limited to date scene domain with scene, but
It can also be used in Novel Temporal Scenario range, so that the utilization scene of speech analysis scheme is no longer single.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
Disclosed attached drawing obtains other attached drawings.
Fig. 1 is a kind of flow chart of speech analysis method disclosed by the embodiments of the present invention；
Fig. 2 is a kind of method flow parsing date value from time related information disclosed by the embodiments of the present invention
Figure；
Fig. 3 is a kind of method flow parsing time numerical value from time related information disclosed by the embodiments of the present invention
Figure；
Fig. 4 is a kind of structural schematic diagram of speech analysis device disclosed by the embodiments of the present invention；
Fig. 5 is a kind of structural schematic diagram of first resolution unit disclosed by the embodiments of the present invention；
Fig. 6 is a kind of structural schematic diagram of second resolution unit disclosed by the embodiments of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a kind of speech analysis method, apparatus and robots, obtain voice input by user and refer to
Enable, in determining the phonetic order comprising after time related information, parsed from the time related information date value and when
Between numerical value obtain a precise time by merging date value and time numerical value.Therefore, compared to traditional scheme and
Speech, the invention enables speech analysis schemes to be no longer limited to date scene domain with scene, but when can also be used in
Between scene domain so that the utilization scene of speech analysis scheme is no longer single.
Referring to Fig. 1, a kind of flow chart of speech analysis method disclosed in one embodiment of the invention, this method is applied to eventually
End, such as robot, including step：
Step S101, phonetic order input by user is obtained；
In practical applications, semantic parsing system can be set in the terminal, it is defeated to user by the semantic parsing system
The phonetic order entered is parsed, and phonetic order input by user can be by the voice command input end that is arranged in terminal
Mouthful, it is input to the semantic parsing system of terminal.
Step S102, judge in phonetic order whether to include time related information, if so, thening follow the steps S103, such as
Fruit is no, then return to step S101；
Time related information described in this step includes：Date information, for example, year, month, day, the beginning of the month, the end of month etc., and,
Temporal information, for example, when, minute, second etc..
In practical applications, regular expression may be used, by the information for including in phonetic order and pre-stored date
Whether information key and temporal information keyword are matched, to determine in phonetic order comprising time related information.
Wherein, when not including time related information in phonetic order, then return to step S101, continues to obtain user's input
Phonetic order.
Step S103, date value is parsed from time related information；
Using default date match algorithm, such as regular expression closes time related information and pre-stored date information
Keyword is matched, when being matched to corresponding date information keyword, the date information keyword that will match to be used as from when
Between the date value that parses in relevant information.
In practical applications, which can be converted to specific year, month, day numerical value.
Step S104, time numerical value is parsed from time related information；
Using preset time matching algorithm, such as regular expression, time related information and pre-stored temporal information are closed
Keyword is matched, when being matched to corresponding temporal information keyword, the temporal information keyword that will match to be used as from when
Between the time numerical value that parses in relevant information.
It should be strongly noted that in practical applications, the execution sequence of step S103 and step S104 include but unlimited
In embodiment shown in FIG. 1, step S104 can also be first carried out, rear to execute step S103, also or two steps are performed simultaneously.
In practical applications, can by the time numerical value conversion be it is specific when, minute, second numerical value.
Step S105, date value and time numerical value are merged, obtains precise time.
By merging date value and time numerical value, one can be obtained while including the accurate of date and time
Time, the precise time can not only be applied to date scene domain, can also apply in Novel Temporal Scenario range.
In summary, speech analysis method disclosed by the invention, obtains phonetic order input by user, is determining the voice
Comprising after time related information in instruction, date value and time numerical value are parsed from the time related information, by by day
Issue value and time numerical value merge, and obtain a precise time.Therefore, for comparing traditional scheme, the invention enables languages
Sound parsing scheme is no longer limited to date scene domain with scene, but can also be used in Novel Temporal Scenario range, to
So that the utilization scene of speech analysis scheme is no longer single.
In order to further optimize the above embodiments, referring to Fig. 2, one kind disclosed in a specific embodiment of the invention is from time correlation
The method flow diagram of date value is parsed in information, the method comprising the steps of：
Step S201, whether time related information is judged comprising the interval date, if so, S202 is thened follow the steps, if
It is no, then follow the steps S203；
The interval date described in this step refers to：The date being spaced for some time apart from current date, for example, today
Afterwards, after some months.
Step S202, the interval date is determined as date value；
That is, when time related information includes the interval date, date for being parsed from time related information
Numerical value is the interval date.
It in practical applications, can be according to date on the same day corresponding standard date format, that is, current date corresponding year
The day moon, and interval date, obtain the date value of interval expiration date on date corresponding standard date format.For example, being currently
On 2 1st, 2018, the interval date was 3 days, then identified date value is on 2 4th, 2018.
Step S203, judge whether time related information includes weekly data, if so, S204 is thened follow the steps, if not,
Then follow the steps S205；
Weekly data described in this step, including：Monday (or Monday), Tuesday (or Tuesday), until Sunday (or star
Phase day) seven weekly datas.
Step S204, weekly data is determined as date value；
That is, when time related information includes weekly data, the day issue that is parsed from time related information
Value is weekly data.
Step S205, judge whether time related information includes the nearest date of current date, if so, thening follow the steps
S206, if not, thening follow the steps S207；
Wherein, the difference of the number of days of the nearest date and the current date is less than threshold value, and the threshold value is according to actual needs
Depending on, such as three days.
The nearest date described in this step refers to apart from the current date closer date, such as today, tomorrow, the day after tomorrow
Deng.
Step S206, the nearest date is determined as date value；
That is, when time related information includes date value, date for being parsed from time related information
Numerical value is " nearest date ".
Step S207, judge whether time related information includes the reference format date, if so, S208 is thened follow the steps,
If not, thening follow the steps S209；
The reference format date described in this step refers to including the date of year, month, day, such as on January 1st, 2018.
Step S208, the reference format date is determined as date value；
That is, when time related information includes reference format date, from time related information parsing obtain
Date value is the reference format date.
Step S209, judgement time related information does not include date value.
In summary, the present invention by judge in order time related information whether comprising interval the date, weekly data, recently
Date or reference format date to determine whether time related information includes date value, and are determining time related information packet
When containing any one being spaced in date, weekly data, nearest date and reference format date, by determining time related information packet
The date data contained is as finally determining date value, and Close Date numerical value judges flow, to realize from time correlation
Parsing obtains date value in information.
In order to further optimize the above embodiments, referring to Fig. 3, one kind disclosed in a specific embodiment of the invention is from time correlation
The method flow diagram of time numerical value is parsed in information, the method comprising the steps of：
Step S301, judge whether time related information includes interval time, if so, S302 is thened follow the steps, if
It is no, then follow the steps S303；
Interval time described in this step refers to：Apart from current time interlude for some time, for example, several small
Shi Hou, after a few minutes.
Step S302, time numerical value will be determined as interval time；
That is, when time related information includes interval time, time for being parsed from time related information
Numerical value is interval time.
Step S303, judge whether time related information includes the reference format time, if so, S304 is thened follow the steps,
If not, thening follow the steps S305；
What the reference format time described in this step referred to includes：The time of Hour Minute Second, for example, 15 points 30 seconds 10 minutes.
Step S304, the reference format time is determined as time numerical value；
That is, when time related information includes reference format time, from time related information parsing obtain
Time numerical value is the reference format time.
Step S305, whether time related information is judged comprising the summary time, if so, S306 is thened follow the steps, if
It is no, then follow the steps S307；
The summary time described in this step refers to：Characterization a period of time, and not comprising when, point and the second time, than
Such as, morning, noon, evening.
Step S306, the summary time it will be determined as time numerical value；
That is, when time related information includes to summarize the time, date for being parsed from time related information
Numerical value is to summarize the time.
Step S307, judge not including time numerical value in time related information.
In summary, when whether the present invention by judging time related information comprising interval time, reference format in order
Between or summarize the time, come determine time related information whether include time numerical value, and determining time related information include interval
Time, reference format time and when summarizing any one in the time, the time data that includes by determining time related information
As finally determining time numerical value, and end time numerical value judges flow, is parsed from time related information to realize
To time numerical value.
To advanced optimize embodiment shown in FIG. 1, step S105 can specifically include：
When time numerical value is earlier than current time, then date value is adjusted to next date value of date value；And
Time numerical value and next date value are merged, precise time is obtained.
Illustrate, it is assumed that parsed from time related information date value be on March 1st, 2018, parse when
Between numerical value be 15 points 20 minutes, the precise time merged is 20 minutes 15 points of on March 1st, 2018, and current time is：2018
On 10 minutes 17 points of March 1, therefore, the precise time parsed from time related information is earlier than current time, in such case
Under, parsing can be obtained the date value in precise time automatically and be automatically adjusted to next date value by terminal, that is, will solution
It analyses obtained precise time and 15 points of March 1 in 2018 is adjusted to 20 minutes 15 points of on March 2nd, 2018 for 20 minutes, so that according to user's language
The precise time that sound is got meets the actual demand of user.
To advanced optimize embodiment shown in FIG. 1, after step S105, can also include：
Precise time is exported, and in the display screen display of the terminal precise time, to facilitate user to check.
Corresponding with above method embodiment, the invention also discloses a kind of speech analysis devices.
Referring to Fig. 4, a kind of structural schematic diagram of speech analysis device, the resolver disclosed in one embodiment of the invention are answered
For terminal, such as robot, which includes：
Acquiring unit 401, for obtaining phonetic order input by user；
In practical applications, semantic parsing system can be set in the terminal, it is defeated to user by the semantic parsing system
The phonetic order entered is parsed, and phonetic order input by user can be by the voice command input end that is arranged in terminal
Mouthful, it is input to the semantic parsing system of terminal.
Judging unit 402, for judging in the phonetic order whether to include time related information；
The time related information includes：Date information, for example, year, month, day, the beginning of the month, the end of month etc., and, time letter
Breath, for example, when, minute, second etc..
In practical applications, regular expression may be used, by the information for including in phonetic order and pre-stored date
Whether information key and temporal information keyword are matched, to determine in phonetic order comprising time related information.
First resolution unit 403, in the case where the judging unit 402 is judged as YES, from the time correlation
Date value is parsed in information；
Using default date match algorithm, such as regular expression closes time related information and pre-stored date information
Keyword is matched, when being matched to corresponding date information keyword, the date information keyword that will match to be used as from when
Between the date value that parses in relevant information.
In practical applications, which can be converted to specific year, month, day numerical value.
Second resolution unit 404, for parsing time numerical value from the time related information；
Using preset time matching algorithm, such as regular expression, time related information and pre-stored temporal information are closed
Keyword is matched, when being matched to corresponding temporal information keyword, the temporal information keyword that will match to be used as from when
Between the time numerical value that parses in relevant information.
It should be noted that the execution sequence of the first resolution unit 403 and the second resolution unit 404 includes but not limited to scheme
4 illustrated embodiments can also first carry out the second resolution unit 404, rear to execute the first resolution unit 403, and also or first parses
Unit 403 and the second resolution unit 404 are performed simultaneously.
Combining unit 405 obtains precise time for merging the date value and the time numerical value.
By merging date value and time numerical value, one can be obtained while including the accurate of date and time
Time, the precise time can not only be applied to date scene domain, can also apply in Novel Temporal Scenario range.
In summary, speech analysis device disclosed by the invention, obtains phonetic order input by user, is determining the voice
Comprising after time related information in instruction, date value and time numerical value are parsed from the time related information, by by day
Issue value and time numerical value merge, and obtain a precise time.Therefore, for comparing traditional scheme, the invention enables languages
Sound parsing scheme is no longer limited to date scene domain with scene, but can also be used in Novel Temporal Scenario range, to
So that the utilization scene of speech analysis scheme is no longer single.
In order to further optimize the above embodiments, referring to Fig. 5, the first parsing of one kind disclosed in a specific embodiment of the invention is single
The structural schematic diagram of member, the first resolution unit can specifically include：
First judgment sub-unit 501, for judging whether the time related information is the interval date；
Wherein, the interval date refers to：The date being spaced for some time apart from current date, for example, after today, it is several
After month.
First determination subelement 502 will be described in the case where first judgment sub-unit 501 is judged as YES
The interval date is determined as the date value；
That is, when time related information includes the interval date, date for being parsed from time related information
Numerical value is the interval date.
It in practical applications, can be according to date on the same day corresponding standard date format, that is, current date corresponding year
The day moon, and interval date, obtain the date value of interval expiration date on date corresponding standard date format.For example, being currently
On 2 1st, 2018, the interval date was 3 days, then identified date value is on 2 4th, 2018.
Second judgment sub-unit 503, in the case where first judgment sub-unit 501 is judged as NO, judging institute
State whether time related information includes weekly data；
Wherein, weekly data, including：Monday (or Monday), Tuesday (or Tuesday), until Sunday (or Sunday) seven
A weekly data.
Second determination subelement 504 will be described in the case where second judgment sub-unit 503 is judged as YES
Weekly data is determined as the date value；
That is, when time related information includes weekly data, the day issue that is parsed from time related information
Value is weekly data.
Third judgment sub-unit 505, in the case where second judgment sub-unit 503 is judged as NO, judging institute
State the nearest date whether time related information includes current date, the difference of the number of days of the nearest date and the current date
Less than threshold value；
Wherein, the difference of the number of days of the nearest date and the current date is less than threshold value, and the threshold value is according to actual needs
Depending on, such as three days.
The nearest date described herein referred to apart from current date closer date, such as today, tomorrow, the day after tomorrow.
Third determination subelement 506 will be described in the case where the third judgment sub-unit 505 is judged as YES
The nearest date is determined as the date value；
That is, when time related information includes date value, date for being parsed from time related information
Numerical value is " nearest date "
4th judgment sub-unit 507, in the case where the third judgment sub-unit 505 is judged as NO, judging institute
State whether time related information includes the reference format date；
Wherein, the reference format date refers to including the date of year, month, day, such as on January 1st, 2018.
4th determination subelement 508 will be described in the case where the 4th judgment sub-unit 507 is judged as YES
The reference format date is determined as the date value；
That is, when time related information includes reference format date, from time related information parsing obtain
Date value is the reference format date.
First judging unit 509, in the case where the 4th judgment sub-unit 508 is judged as NO, described in judgement
Time related information does not include the date value.
In summary, the present invention by judge in order time related information whether comprising interval the date, weekly data, recently
Date or reference format date to determine whether time related information includes date value, and are determining time related information packet
When containing any one being spaced in date, weekly data, nearest date and reference format date, by determining time related information packet
The date data contained is as finally determining date value, and Close Date numerical value judges flow, to realize from time correlation
Parsing obtains date value in information.
In order to further optimize the above embodiments, referring to Fig. 6, the second parsing of one kind disclosed in a specific embodiment of the invention is single
The structural schematic diagram of member, the second resolution unit can specifically include：
5th judgment sub-unit 601, for judging whether the time related information includes interval time；
Wherein, interval time refers to：Apart from current time interlude for some time, for example, after a few houres, it is several
After minute.
5th determination subelement 602 will be described in the case where the 5th judgment sub-unit 601 is judged as YES
Interval time is determined as the time numerical value；
That is, when time related information includes interval time, time for being parsed from time related information
Numerical value is interval time.
6th judgment sub-unit 603, in the case where the 5th judgment sub-unit 601 is judged as NO, judging institute
State whether time related information includes the reference format time；
Wherein, what the reference format time referred to includes：The time of Hour Minute Second, for example, 15 points 30 seconds 10 minutes.
6th determination subelement 604 will be described in the case where the 6th judgment sub-unit 603 is judged as YES
The reference format time is determined as the time numerical value；
That is, when time related information includes reference format time, from time related information parsing obtain
Time numerical value is the reference format time.
7th judgment sub-unit 605, in the case where the 6th judgment sub-unit 603 is judged as NO, judging institute
Time related information is stated whether comprising summarizing the time, summarys time is characterization a period of time, and not comprising when, divide and the second
Time, for example, morning, noon, evening.
7th determination subelement 606 will be described in the case where the 7th judgment sub-unit 605 is judged as YES
The summary time is determined as the time numerical value；
That is, when time related information includes to summarize the time, date for being parsed from time related information
Numerical value is to summarize the time.
Second judgement subelement 607, in the case where the 7th judgment sub-unit 605 is judged as NO, judging institute
It states and does not include the time numerical value in time related information.
In summary, when whether the present invention by judging time related information comprising interval time, reference format in order
Between or summarize the time, come determine time related information whether include time numerical value, and determining time related information include interval
Time, reference format time and when summarizing any one in the time, the time data that includes by determining time related information
As finally determining time numerical value, and end time numerical value judges flow, is parsed from time related information to realize
To time numerical value.
To advanced optimize embodiment shown in Fig. 4, combining unit 405 can specifically include：
Subelement is adjusted, for working as the time numerical value earlier than current time, then the date value is adjusted to described
Next date value of date value；
Merge subelement and obtains the essence for merging the time numerical value and next date value
The true time.
Illustrate, it is assumed that parsed from time related information date value be on March 1st, 2018, parse when
Between numerical value be 15 points 20 minutes, the precise time merged is 20 minutes 15 points of on March 1st, 2018, and current time is：2018
On 10 minutes 17 points of March 1, therefore, the precise time parsed from time related information is earlier than current time, in such case
Under, parsing can be obtained the date value in precise time automatically and be automatically adjusted to next date value by terminal, that is, will solution
It analyses obtained precise time and 15 points of March 1 in 2018 is adjusted to 20 minutes 15 points of on March 2nd, 2018 for 20 minutes, so that according to user's language
The precise time that sound is got meets the actual demand of user.
To advanced optimize embodiment shown in Fig. 4, speech analysis device can also include：
Output unit, for exporting the precise time.
It in practical applications, can also be by the precise time in the display screen display of the terminal precise time, to facilitate use
Family is checked.
The invention also discloses a kind of robot, which includes the speech analysis device in above-described embodiment, machine
People refers to above-described embodiment corresponding part, details are not described herein again to the resolving for phonetic order.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only that
A little elements, but also include other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with other
The difference of embodiment, just to refer each other for identical similar portion between each embodiment.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention.
Various modifications to these embodiments will be apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest range caused.
Claims (11)
1. a kind of speech analysis method, which is characterized in that including：
Obtain phonetic order input by user；
Judge in the phonetic order whether to include time related information；
If it is, parsing date value from the time related information；
Time numerical value is parsed from the time related information；
The date value and the time numerical value are merged, precise time is obtained.
2. speech analysis method according to claim 1, which is characterized in that described to be parsed from the time related information
The process for going out date value, specifically includes：
Judge whether the time related information is the interval date；
If the time related information includes the interval date, the interval date is determined as the date value；
If the time related information does not include the interval date, judge the time related information whether comprising all numbers
According to；
If the time related information includes the weekly data, the weekly data is determined as the date value；
If the time related information does not include the weekly data, judge whether the time related information includes to work as the day before yesterday
The difference of the number of days of the nearest date of phase, the nearest date and the current date is less than threshold value；
If the time related information includes the nearest date, the nearest date is determined as the date value；
If the time related information does not include the nearest date, judge whether the time related information includes standard
The format date；
If the time correlation date includes the reference format date, the reference format date is determined as the day
Issue value；
If the time correlation date does not include the reference format date, judge that the time related information does not include institute
State date value.
3. speech analysis method according to claim 1, which is characterized in that described to be parsed from the time related information
The process for going out time numerical value, specifically includes：
Judge whether the time related information includes interval time；
If the time related information includes the interval time, the interval time is determined as the time numerical value；
If the time related information does not include the interval time, judge whether the time related information includes standard
The format time；
If the time related information includes the reference format time, when the reference format time is determined as described
Between numerical value；
If the time related information do not include the reference format time, judge the time related information whether include
Summarize the time, the summarys time be characterization a period of time, and not comprising when, divide and the second time；
If the time related information includes the summary time, the summary time is determined as the time numerical value；
If the time related information does not include the summary time, judge not including in the time related information described
Time numerical value.
4. speech analysis method according to claim 1, which is characterized in that by the date value and the time number
Value merges, and obtains precise time, specifically includes：
When the time numerical value is earlier than current time, then the date value is adjusted to next date of the date value
Numerical value；
The time numerical value and next date value are merged, the precise time is obtained.
5. speech analysis method according to claim 1, which is characterized in that further include：
Export the precise time.
6. a kind of speech analysis device, which is characterized in that including：
Acquiring unit, for obtaining phonetic order input by user；
Judging unit, for judging in the phonetic order whether to include time related information；
First resolution unit, in the case where the judging unit is judged as YES, being parsed from the time related information
Go out date value；
Second resolution unit, for parsing time numerical value from the time related information；
Combining unit obtains precise time for merging the date value and the time numerical value.
7. speech analysis device according to claim 6, which is characterized in that first resolution unit specifically includes：
First judgment sub-unit, for judging whether the time related information is the interval date；
First determination subelement, it is in the case where first judgment sub-unit is judged as YES, the interval date is true
It is set to the date value；
Second judgment sub-unit, in the case where first judgment sub-unit is judged as NO, judging the time correlation
Whether information includes weekly data；
Second determination subelement, in the case where second judgment sub-unit is judged as YES, the weekly data to be determined
For the date value；
Third judgment sub-unit, in the case where second judgment sub-unit is judged as NO, judging the time correlation
Whether information includes the nearest date of current date, and the difference of the number of days of the nearest date and the current date is less than threshold value；
Third determination subelement, it is in the case where the third judgment sub-unit is judged as YES, the nearest date is true
It is set to the date value；
4th judgment sub-unit, in the case where the third judgment sub-unit is judged as NO, judging the time correlation
Whether information includes the reference format date；
4th determination subelement, in the case where the 4th judgment sub-unit is judged as YES, by the reference format day
Phase is determined as the date value；
First judging unit, in the case where the 4th judgment sub-unit is judged as NO, judging the time correlation letter
Breath does not include the date value.
8. speech analysis device according to claim 6, which is characterized in that second resolution unit specifically includes：
5th judgment sub-unit, for judging whether the time related information includes interval time；
5th determination subelement, it is in the case where the 5th judgment sub-unit is judged as YES, the interval time is true
It is set to the time numerical value；
6th judgment sub-unit, in the case where the 5th judgment sub-unit is judged as NO, judging the time correlation
Whether information includes the reference format time；
6th determination subelement, in the case where the 6th judgment sub-unit is judged as YES, when by the reference format
Between be determined as the time numerical value；
7th judgment sub-unit, in the case where the 6th judgment sub-unit is judged as NO, judging the time correlation
Information whether comprising summarize the time, the summarys time be characterization a period of time, and not comprising when, divide and the second time；
7th determination subelement, it is in the case where the 7th judgment sub-unit is judged as YES, the summary time is true
It is set to the time numerical value；
Second judgement subelement, in the case where the 7th judgment sub-unit is judged as NO, judging the time correlation
The time numerical value is not included in information.
9. speech analysis device according to claim 6, which is characterized in that the combining unit specifically includes：
Subelement is adjusted, for working as the time numerical value earlier than current time, then the date value is adjusted to the date
Next date value of numerical value；
Merge subelement, for merging the time numerical value and next date value, when obtaining described accurate
Between.
10. speech analysis device according to claim 6, which is characterized in that further include：
Output unit, for exporting the precise time.
11. a kind of robot, which is characterized in that including the speech analysis device described in claim 6~10 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810300973.5A CN108597504A (en) | 2018-04-04 | 2018-04-04 | A kind of speech analysis method, apparatus and robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810300973.5A CN108597504A (en) | 2018-04-04 | 2018-04-04 | A kind of speech analysis method, apparatus and robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108597504A true CN108597504A (en) | 2018-09-28 |
Family
ID=63624498
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810300973.5A Pending CN108597504A (en) | 2018-04-04 | 2018-04-04 | A kind of speech analysis method, apparatus and robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108597504A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004095422A3 (en) * | 2003-04-22 | 2005-05-12 | Spinvox Ltd | Operator performed voicemall transcription |
| CN1809188A (en) * | 2005-01-18 | 2006-07-26 | 乐金电子(中国)研究开发中心有限公司 | Voice calendar managing device and method of mobile communication terminal |
| CN104021787A (en) * | 2014-06-13 | 2014-09-03 | 中国民航信息网络股份有限公司 | Air ticket searching system and method based on voice recognition |
| CN204765586U (en) * | 2015-06-30 | 2015-11-18 | 深圳欧德蒙科技有限公司 | Wrist formula pulse record and analytical equipment |
| CN106020953A (en) * | 2016-05-12 | 2016-10-12 | 青岛海信移动通信技术股份有限公司 | Method and device for establishing schedule in electronic calendar |
| CN205943036U (en) * | 2016-08-22 | 2017-02-08 | 河源市勇艺达科技股份有限公司 | System is reminded by robot based on pronunciation set up |
| CN106409295A (en) * | 2015-07-31 | 2017-02-15 | 腾讯科技（深圳）有限公司 | Method and apparatus for recognizing time information from natural voice information |
| CN106776537A (en) * | 2016-11-18 | 2017-05-31 | 畅捷通信息技术股份有限公司 | The abstracting method and system of temporal information and subject information in text |
| CN106881721A (en) * | 2016-09-21 | 2017-06-23 | 摩瑞尔电器（昆山）有限公司 | The purchase method and system of service robot |
-
2018
- 2018-04-04 CN CN201810300973.5A patent/CN108597504A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004095422A3 (en) * | 2003-04-22 | 2005-05-12 | Spinvox Ltd | Operator performed voicemall transcription |
| CN1809188A (en) * | 2005-01-18 | 2006-07-26 | 乐金电子(中国)研究开发中心有限公司 | Voice calendar managing device and method of mobile communication terminal |
| CN104021787A (en) * | 2014-06-13 | 2014-09-03 | 中国民航信息网络股份有限公司 | Air ticket searching system and method based on voice recognition |
| CN204765586U (en) * | 2015-06-30 | 2015-11-18 | 深圳欧德蒙科技有限公司 | Wrist formula pulse record and analytical equipment |
| CN106409295A (en) * | 2015-07-31 | 2017-02-15 | 腾讯科技（深圳）有限公司 | Method and apparatus for recognizing time information from natural voice information |
| CN106020953A (en) * | 2016-05-12 | 2016-10-12 | 青岛海信移动通信技术股份有限公司 | Method and device for establishing schedule in electronic calendar |
| CN205943036U (en) * | 2016-08-22 | 2017-02-08 | 河源市勇艺达科技股份有限公司 | System is reminded by robot based on pronunciation set up |
| CN106881721A (en) * | 2016-09-21 | 2017-06-23 | 摩瑞尔电器（昆山）有限公司 | The purchase method and system of service robot |
| CN106776537A (en) * | 2016-11-18 | 2017-05-31 | 畅捷通信息技术股份有限公司 | The abstracting method and system of temporal information and subject information in text |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102708100B (en) | Method and device for digging relation keyword of relevant entity word and application thereof | |
| WO2019093239A1 (en) | Information processing device, method, and recording medium | |
| TWI711967B (en) | Method, device and equipment for determining broadcast voice | |
| WO2017020657A1 (en) | Method and device for identifying time information from voice information | |
| CN103268311A (en) | Event-structure-based Chinese statement analysis method | |
| WO2021254046A1 (en) | Information query method and device | |
| CN109637536B (en) | Method and device for automatically identifying semantic accuracy | |
| WO2020230383A1 (en) | Automatic report creation system | |
| CN105117107A (en) | Application program icon managing method and application program icon managing system | |
| AU2002344272A1 (en) | Method and system in an office application for providing content dependent help information | |
| Rohrdantz et al. | Lexical Semantics and Distribution of Suffixes-A Visual Analysis | |
| JP2007272440A (en) | Behavior advice providing system and behavior advice providing device | |
| CN106055641B (en) | Intelligent robot-oriented man-machine interaction method and device | |
| WO2013010477A1 (en) | Method and device for text reminder | |
| CN107894978B (en) | Time word extraction method and device | |
| KR20190107682A (en) | Systems and methods for processing input streams of calendar applications | |
| CN108597504A (en) | A kind of speech analysis method, apparatus and robot | |
| CN103761096B (en) | The automated input methodology realized based on IBinder mechanism and device | |
| Pasma | Linguistic metaphor identification in Dutch | |
| CN112002083B (en) | Business blockage early warning system applied to bank self-service equipment | |
| Cisco | Formulas | |
| CN113157966A (en) | Display method and device and electronic equipment | |
| CN108281144A (en) | A kind of audio recognition method and system | |
| CN111177418A (en) | Method and device for acquiring time text and storage medium | |
| US20130332828A1 (en) | Auxiliary method and device in information configuration process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180928 |