CN102542705A - Voice reminding method and system - Google Patents
Voice reminding method and system Download PDFInfo
- Publication number
- CN102542705A CN102542705A CN2010106220609A CN201010622060A CN102542705A CN 102542705 A CN102542705 A CN 102542705A CN 2010106220609 A CN2010106220609 A CN 2010106220609A CN 201010622060 A CN201010622060 A CN 201010622060A CN 102542705 A CN102542705 A CN 102542705A
- Authority
- CN
- China
- Prior art keywords
- voice
- content
- speech
- user
- memory storage
- 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
Images
Landscapes
- Machine Translation (AREA)
Abstract
The invention discloses a voice reminding method and system. by the method and the system, a user can conveniently input reminding and the direct feeling of a reminded user is enhanced. According to the technical scheme, the method comprises the following steps: receiving a voice input of the user; identifying and storing content information in the voice input according to the voice input; and reminding the user according to the identified content information, wherein the reminded content is the stored voice content.
Description
Technical field
The present invention relates to the technology of voice reminder, relate in particular to the method and system that the car owner is carried out voice reminder through car-mounted terminal.
Background technology
The function (or being called prompting function) that p.m.entry is arranged on some electronic equipments usually, the user imports the time point that needs the information of reminding and prompting is set.When the time point that is provided with arrived, electronic equipment can remind the user to have reminder events to take place through certain alerting pattern (for example quarter-bell), and concrete content can show on the screen of electronic equipment.
Inconvenience below this alerting pattern exists:
1, this mode needs the time point that the user reminds in the software Chinese words input reminded contents and the selection of electronic equipment, and input mode is loaded down with trivial details.If applied environment is in the driving process of vehicle, then the car owner reminds the problem that can bring on the traffic safety that is provided with.
2, the alerting pattern of this mode is direct inadequately, and when reminding quarter-bell to open, the user can't directly be known the content of prompting, and need press corresponding button, enters into the content that current prompting clauses and subclauses are checked prompting.Same, if applied environment is in the driving process of vehicle, then the car owner gets into annoyware and checks that reminded contents also can bring the problem on the traffic safety.
Summary of the invention
The objective of the invention is to address the above problem, a kind of method of voice reminder is provided, made things convenient for the user to import the mode of prompting, strengthened the direct feel of reminding the user.
Another object of the present invention is to provide a kind of system of voice reminder.
Technical scheme of the present invention is: the present invention has disclosed a kind of method of voice reminder, comprising:
The input of reception user's voice;
According to phonetic entry identification wherein content information and store;
Content information according to identifying is reminded, the voice content of the content of prompting for having stored.
According to an embodiment of the method for voice reminder of the present invention, receive phonetic entry, the identification content information also stores, reminds the voice content of having stored all to accomplish at car-mounted terminal.
According to an embodiment of the method for voice reminder of the present invention, the content information that is stored in the phonetic entry of car-mounted terminal exports to computer end.
According to an embodiment of the method for voice reminder of the present invention, after receiving the user's voice input and before carrying out content recognition, also comprise noise reduction process is carried out in phonetic entry according to phonetic entry.
According to an embodiment of the method for voice reminder of the present invention, the mode of prompting is that the mode that adopts the loudspeaker of car-mounted terminal to carry out voice playing realizes.
The present invention has also disclosed a kind of system of voice reminder, comprising:
Speech input device receives the voice data that the user imports;
Speech engine couples speech input device, the content information that the recognizing voice input is comprised;
Memory storage couples speech engine, the content information that phonetic entry comprised of storaged voice engine output;
Playing device couples this memory storage, reminds according to the content information that identifies, and the content of prompting is the voice content of having stored in the memory storage.
According to an embodiment of the system of voice reminder of the present invention, speech input device, speech engine, memory storage, playing device are integrated in car-mounted terminal.
According to an embodiment of the system of voice reminder of the present invention, this system also comprises:
Data transmission interface couples memory storage, and the data in the memory storage is transferred to external unit.
According to an embodiment of the system of voice reminder of the present invention, this system also comprises:
Denoising device couples speech input device and speech engine, and noise reduction process is carried out in phonetic entry.
The present invention contrasts prior art has following beneficial effect: technical scheme of the present invention is to receive the user's voice input earlier; Again according to phonetic entry identification wherein content information and store; The content information that last basis identifies is reminded, and the content of prompting is the voice content of having stored.The contrast prior art, one aspect of the present invention has substituted traditional literal input with phonetic entry, has substituted traditional prompting that needs the user to browse with voice reminder on the one hand.
Description of drawings
The process flow diagram of first embodiment of the method that shows voice reminder of the present invention that Fig. 1 is exemplary.
The process flow diagram of second embodiment of the method that shows voice reminder of the present invention that Fig. 2 is exemplary.
The process flow diagram of the 3rd embodiment of the method that shows voice reminder of the present invention that Fig. 3 is exemplary.
The schematic diagram of first embodiment of the system that shows voice reminder of the present invention that Fig. 4 is exemplary.
The schematic diagram of second embodiment of the system that shows voice reminder of the present invention that Fig. 5 is exemplary.
The schematic diagram of the 3rd embodiment of the system that shows voice reminder of the present invention that Fig. 6 is exemplary.
Embodiment
Below in conjunction with accompanying drawing and embodiment the present invention is done further description.
First embodiment of the method for voice reminder
Fig. 1 shows the flow process of first embodiment of the method for voice reminder of the present invention.See also Fig. 1, the detailed step of the method for the voice reminder of present embodiment details as follows.
Step S10: receive the user's voice input.
At vehicle-mounted end, the user is through the input of the audio input interface on vehicle-mounted end voice, and for example, the user says " this afternoon, 3 promptings had meeting ", and mobile unit receives these input voice of user.
Step S12: the content information in the recognizing voice input is also stored.
At vehicle-mounted end a speech engine is installed, speech engine receives user's input voice, the content that identifies in the voice to be comprised, with these content stores at vehicle-mounted end.
The speech recognition technology of speech engine is existing technology.For example; Speech engine comprises voice acquisition device, front-end processing module, characteristic extracting module, fundamental tone characteristic extracting module, Tone recognition module, training module, acoustic layer identification module, phonetic syntax Understanding Module, nine parts of language understanding module; This acoustic layer identification module adopts the hidden Markov model of representing with segment length's probability of state; The hidden Markov model that promptly distributes based on the segment length; Be called for short the DDBHMM model, the segment length of this model obeys the distribution with convexity, wherein; Voice units such as phoneme in the state in this model and the voice or syllable are corresponding, and the phonic signal character of these voice unit pronunciations is as the observed quantity of corresponding voice unit; The training method step of this training module is following: to the one or more pronunciation sample files that include some specific word of training module input; The proper vector of the series of frames of every words in this document is unit with the frame; Through the training searching algorithm in the training module; Each state to each speech in every frame voice signal is searched for comparison, obtains the Codebook of Vector Quantization (VQ code book) of phonic signal character vector and the DDBHMM model parameter of this specific word, inputs to the acoustic layer identification module; The audio recognition method step of described Chinese continuous speech recognition system is following: the voice signal through voice acquisition device reception people, carry out front-end processing for the voice signal of importing, and carry out the MFCC phonetic feature (based on the phonetic feature of Mel cepstrum coefficient; Mel-Frequency Cepstral Coefficients) extraction of sequence, this MFCC phonetic feature sequence that obtains is admitted to the acoustic layer identification module, through the searching algorithm of acoustic layer identification module; Produce the recognition result of pinyin lattice form, simultaneously, the fundamental tone eigenvector of voice signal also is extracted out; Send into the Tone recognition module, the Tone recognition module is utilized the breakpoint information of fundamental tone characteristic information and phonetic, obtains the tone information of phonetic and joins in the pinyin lattice; Then; Through the phonetic syntax Understanding Module pinyin lattice is pruned, the Syllable Lattice after simplifying is admitted to the language understanding module, is converted into phonetic figure and speech figure; And in speech figure, search for, get the result that understands to the end.
For example, speech engine can identify the particular content of " this afternoon, 3 promptings had meeting ", knows that this is a reminder events, and the content of prompting is " this afternoon 3 meeting is arranged ".
Step S14: the content information according to identifying is reminded, the voice content of the content of prompting for having stored.
To have identified this be a reminder events to vehicle-mounted end in a last step, and be to remind 3 of this afternoons meeting to be arranged.Therefore, in 3 the moment of this afternoon, vehicle-mounted end triggers a reminder events, and informs the user through the mode of voice reminder, i.e. content through the audio-frequence player device on the vehicle-mounted end (for example loudspeaker) broadcast " this afternoon 3 meeting is arranged ".
Can find out that from this embodiment the voice content that reception phonetic entry, identification content information, content information stored, prompting have been stored is all accomplished at car-mounted terminal.
Second embodiment of the method for voice reminder
Fig. 2 shows the flow process of second embodiment of the method for voice reminder of the present invention.See also Fig. 2, the detailed step of the method for the voice reminder of present embodiment details as follows.
Step S20: receive the user's voice input.
At vehicle-mounted end, the user is through the input of the audio input interface on vehicle-mounted end voice, and for example, the user says " this afternoon, 3 promptings had meeting ", and mobile unit receives these input voice of user.
Step S22: the content information in the recognizing voice input is also stored.
At vehicle-mounted end a speech engine is installed, speech engine receives user's input voice, the content that identifies in the voice to be comprised, with these content stores at vehicle-mounted end.
The speech recognition technology of speech engine is existing technology.For example; Speech engine comprises voice acquisition device, front-end processing module, characteristic extracting module, fundamental tone characteristic extracting module, Tone recognition module, training module, acoustic layer identification module, phonetic syntax Understanding Module, nine parts of language understanding module; This acoustic layer identification module adopts the hidden Markov model of representing with segment length's probability of state; The hidden Markov model that promptly distributes based on the segment length; Be called for short the DDBHMM model, the segment length of this model obeys the distribution with convexity, wherein; Voice units such as phoneme in the state in this model and the voice or syllable are corresponding, and the phonic signal character of these voice unit pronunciations is as the observed quantity of corresponding voice unit; The training method step of this training module is following: to the one or more pronunciation sample files that include some specific word of training module input; The proper vector of the series of frames of every words in this document is unit with the frame; Through the training searching algorithm in the training module; Each state to each speech in every frame voice signal is searched for comparison, obtains the Codebook of Vector Quantization (VQ code book) of phonic signal character vector and the DDBHMM model parameter of this specific word, inputs to the acoustic layer identification module; The audio recognition method step of described Chinese continuous speech recognition system is following: the voice signal through voice acquisition device reception people, carry out front-end processing for the voice signal of importing, and carry out the MFCC phonetic feature (based on the phonetic feature of Mel cepstrum coefficient; Mel-Frequency Cepstral Coefficients) extraction of sequence, this MFCC phonetic feature sequence that obtains is admitted to the acoustic layer identification module, through the searching algorithm of acoustic layer identification module; Produce the recognition result of pinyin lattice form, simultaneously, the fundamental tone eigenvector of voice signal also is extracted out; Send into the Tone recognition module, the Tone recognition module is utilized the breakpoint information of fundamental tone characteristic information and phonetic, obtains the tone information of phonetic and joins in the pinyin lattice; Then; Through the phonetic syntax Understanding Module pinyin lattice is pruned, the Syllable Lattice after simplifying is admitted to the language understanding module, is converted into phonetic figure and speech figure; And in speech figure, search for, get the result that understands to the end.
For example, speech engine can identify the particular content of " this afternoon, 3 promptings had meeting ", knows that this is a reminder events, and the content of prompting is " this afternoon 3 meeting is arranged ".
Step S24: the content information according to identifying is reminded, the voice content of the content of prompting for having stored.
To have identified this be a reminder events to vehicle-mounted end in a last step, and be to remind 3 of this afternoons meeting to be arranged.Therefore, in 3 the moment of this afternoon, vehicle-mounted end triggers a reminder events, and informs the user through the mode of voice reminder, i.e. content through the audio-frequence player device on the vehicle-mounted end (for example loudspeaker) broadcast " this afternoon 3 meeting is arranged ".
Can find out that from this embodiment the voice content that reception phonetic entry, identification content information, content information stored, prompting have been stored is all accomplished at car-mounted terminal.
Step S26: the content information that will be stored in the phonetic entry of car-mounted terminal exports to computer.
Offer the function that the user backs up on computers and edits.
The 3rd embodiment of the method for voice reminder
Fig. 3 shows the flow process of the 3rd embodiment of the method for voice reminder of the present invention.See also Fig. 3, the detailed step of the method for the voice reminder of present embodiment details as follows.
Step S30: receive the user's voice input.
At vehicle-mounted end, the user is through the input of the audio input interface on vehicle-mounted end voice, and for example, the user says " this afternoon, 3 promptings had meeting ", and mobile unit receives these input voice of user.
Step S32: noise reduction process is carried out in phonetic entry.
Step S34: the content information in the recognizing voice input is also stored.
At vehicle-mounted end a speech engine is installed, speech engine receives user's input voice, the content that identifies in the voice to be comprised, with these content stores at vehicle-mounted end.
The speech recognition technology of speech engine is existing technology.For example; Speech engine comprises voice acquisition device, front-end processing module, characteristic extracting module, fundamental tone characteristic extracting module, Tone recognition module, training module, acoustic layer identification module, phonetic syntax Understanding Module, nine parts of language understanding module; This acoustic layer identification module adopts the hidden Markov model of representing with segment length's probability of state; The hidden Markov model that promptly distributes based on the segment length; Be called for short the DDBHMM model, the segment length of this model obeys the distribution with convexity, wherein; Voice units such as phoneme in the state in this model and the voice or syllable are corresponding, and the phonic signal character of these voice unit pronunciations is as the observed quantity of corresponding voice unit; The training method step of this training module is following: to the one or more pronunciation sample files that include some specific word of training module input; The proper vector of the series of frames of every words in this document is unit with the frame; Through the training searching algorithm in the training module; Each state to each speech in every frame voice signal is searched for comparison, obtains the Codebook of Vector Quantization (VQ code book) of phonic signal character vector and the DDBHMM model parameter of this specific word, inputs to the acoustic layer identification module; The audio recognition method step of described Chinese continuous speech recognition system is following: the voice signal through voice acquisition device reception people, carry out front-end processing for the voice signal of importing, and carry out the MFCC phonetic feature (based on the phonetic feature of Mel cepstrum coefficient; Mel-Frequency Cepstral Coefficients) extraction of sequence, this MFCC phonetic feature sequence that obtains is admitted to the acoustic layer identification module, through the searching algorithm of acoustic layer identification module; Produce the recognition result of pinyin lattice form, simultaneously, the fundamental tone eigenvector of voice signal also is extracted out; Send into the Tone recognition module, the Tone recognition module is utilized the breakpoint information of fundamental tone characteristic information and phonetic, obtains the tone information of phonetic and joins in the pinyin lattice; Then; Through the phonetic syntax Understanding Module pinyin lattice is pruned, the Syllable Lattice after simplifying is admitted to the language understanding module, is converted into phonetic figure and speech figure; And in speech figure, search for, get the result that understands to the end.
For example, speech engine can identify the particular content of " this afternoon, 3 promptings had meeting ", knows that this is a reminder events, and the content of prompting is " this afternoon 3 meeting is arranged ".
Step S36: the content information according to identifying is reminded, the voice content of the content of prompting for having stored.
To have identified this be a reminder events to vehicle-mounted end in a last step, and be to remind 3 of this afternoons meeting to be arranged.Therefore, in 3 the moment of this afternoon, vehicle-mounted end triggers a reminder events, and informs the user through the mode of voice reminder, i.e. content through the audio-frequence player device on the vehicle-mounted end (for example loudspeaker) broadcast " this afternoon 3 meeting is arranged ".
Can find out that from this embodiment the voice content that reception phonetic entry, identification content information, content information stored, prompting have been stored is all accomplished at car-mounted terminal.
First embodiment of the system of voice reminder
Fig. 4 shows the principle of first embodiment of the system of voice reminder of the present invention.See also Fig. 4, the system of the voice reminder of present embodiment comprises: speech input device 10, speech engine 12, memory storage 14, playing device 16.
Annexation between these devices is: speech input device 10 couples speech engine 12, and speech engine 12 couples memory storage 14, and memory storage 14 couples playing device 16.
The operation logic of the system of the voice reminder of present embodiment is following.
At vehicle-mounted end, an example of speech input device 10 is the audio input interfaces on the car-mounted terminal.The user is through the input of the audio input interface on vehicle-mounted end voice, and for example, the user says " this afternoon, 3 promptings had meeting ", and mobile unit receives these input voice of user.
At vehicle-mounted end a speech engine 12 is installed, speech engine 12 receives users' input voice, the content that identifies in the voice to be comprised, with these content stores in the memory storage 14 of vehicle-mounted end.
The speech recognition technology of speech engine 12 is existing technology.For example; Speech engine 12 comprises voice acquisition device, front-end processing module, characteristic extracting module, fundamental tone characteristic extracting module, Tone recognition module, training module, acoustic layer identification module, phonetic syntax Understanding Module, nine parts of language understanding module; This acoustic layer identification module adopts the hidden Markov model of representing with segment length's probability of state; The hidden Markov model that promptly distributes based on the segment length; Be called for short the DDBHMM model, the segment length of this model obeys the distribution with convexity, wherein; Voice units such as phoneme in the state in this model and the voice or syllable are corresponding, and the phonic signal character of these voice unit pronunciations is as the observed quantity of corresponding voice unit; The training method step of this training module is following: to the one or more pronunciation sample files that include some specific word of training module input; The proper vector of the series of frames of every words in this document is unit with the frame; Through the training searching algorithm in the training module; Each state to each speech in every frame voice signal is searched for comparison, obtains the Codebook of Vector Quantization (VQ code book) of phonic signal character vector and the DDBHMM model parameter of this specific word, inputs to the acoustic layer identification module; The audio recognition method step of described Chinese continuous speech recognition system is following: the voice signal through voice acquisition device reception people, carry out front-end processing for the voice signal of importing, and carry out the MFCC phonetic feature (based on the phonetic feature of Mel cepstrum coefficient; Mel-Frequency Cepstral Coefficients) extraction of sequence, this MFCC phonetic feature sequence that obtains is admitted to the acoustic layer identification module, through the searching algorithm of acoustic layer identification module; Produce the recognition result of pinyin lattice form, simultaneously, the fundamental tone eigenvector of voice signal also is extracted out; Send into the Tone recognition module, the Tone recognition module is utilized the breakpoint information of fundamental tone characteristic information and phonetic, obtains the tone information of phonetic and joins in the pinyin lattice; Then; Through the phonetic syntax Understanding Module pinyin lattice is pruned, the Syllable Lattice after simplifying is admitted to the language understanding module, is converted into phonetic figure and speech figure; And in speech figure, search for, get the result that understands to the end.
For example, speech engine 12 can identify the particular content of " this afternoon, 3 promptings had meeting ", knows that this is a reminder events, and the content of prompting is " this afternoon 3 meeting is arranged ".
Playing device 16 is reminded according to the content information that identifies, the voice content of the content of prompting for having stored.
Having identified this at the speech engine 12 of vehicle-mounted end is a reminder events, and is to remind 3 of this afternoons meeting to be arranged.Therefore, in 3 the moment of this afternoon, vehicle-mounted end triggers a reminder events, and informs the user through the mode of voice reminder, i.e. content through the playing device on the vehicle-mounted end 16 (for example loudspeaker) broadcast " this afternoon 3 meeting is arranged ".
Can find out that from this embodiment speech input device 10, speech engine 12, memory storage 14 and playing device 16 all are integrated on the car-mounted terminal.
Second embodiment of the system of voice reminder
Fig. 5 shows the principle of second embodiment of the system of voice reminder of the present invention.See also Fig. 5, the system of the voice reminder of present embodiment comprises: speech input device 20, speech engine 22, memory storage 24, playing device 26 and data transmission interface 28.
Annexation between these devices is: speech input device 20 couples speech engine 22, and speech engine 22 couples memory storage 24, and memory storage 24 couples playing device 26, and memory storage 24 couples data transmission interface 28.
The operation logic of the system of the voice reminder of present embodiment is following.
Speech input device 20 receives the user's voice input.
At vehicle-mounted end, an example of speech input device 20 is the audio input interfaces on the car-mounted terminal.The user is through the input of the audio input interface on vehicle-mounted end voice, and for example, the user says " this afternoon, 3 promptings had meeting ", and mobile unit receives these input voice of user.
At vehicle-mounted end a speech engine 22 is installed, speech engine 22 receives users' input voice, the content that identifies in the voice to be comprised, with these content stores in the memory storage 24 of vehicle-mounted end.
The speech recognition technology of speech engine 22 is existing technology.For example; Speech engine 22 comprises voice acquisition device, front-end processing module, characteristic extracting module, fundamental tone characteristic extracting module, Tone recognition module, training module, acoustic layer identification module, phonetic syntax Understanding Module, nine parts of language understanding module; This acoustic layer identification module adopts the hidden Markov model of representing with segment length's probability of state; The hidden Markov model that promptly distributes based on the segment length; Be called for short the DDBHMM model, the segment length of this model obeys the distribution with convexity, wherein; Voice units such as phoneme in the state in this model and the voice or syllable are corresponding, and the phonic signal character of these voice unit pronunciations is as the observed quantity of corresponding voice unit; The training method step of this training module is following: to the one or more pronunciation sample files that include some specific word of training module input; The proper vector of the series of frames of every words in this document is unit with the frame; Through the training searching algorithm in the training module; Each state to each speech in every frame voice signal is searched for comparison, obtains the Codebook of Vector Quantization (VQ code book) of phonic signal character vector and the DDBHMM model parameter of this specific word, inputs to the acoustic layer identification module; The audio recognition method step of described Chinese continuous speech recognition system is following: the voice signal through voice acquisition device reception people, carry out front-end processing for the voice signal of importing, and carry out the MFCC phonetic feature (based on the phonetic feature of Mel cepstrum coefficient; Mel-Frequency Cepstral Coefficients) extraction of sequence, this MFCC phonetic feature sequence that obtains is admitted to the acoustic layer identification module, through the searching algorithm of acoustic layer identification module; Produce the recognition result of pinyin lattice form, simultaneously, the fundamental tone eigenvector of voice signal also is extracted out; Send into the Tone recognition module, the Tone recognition module is utilized the breakpoint information of fundamental tone characteristic information and phonetic, obtains the tone information of phonetic and joins in the pinyin lattice; Then; Through the phonetic syntax Understanding Module pinyin lattice is pruned, the Syllable Lattice after simplifying is admitted to the language understanding module, is converted into phonetic figure and speech figure; And in speech figure, search for, get the result that understands to the end.
For example, speech engine 22 can identify the particular content of " this afternoon, 3 promptings had meeting ", knows that this is a reminder events, and the content of prompting is " this afternoon 3 meeting is arranged ".
Playing device 26 is reminded according to the content information that identifies, the voice content of the content of prompting for having stored.
Having identified this at the speech engine 22 of vehicle-mounted end is a reminder events, and is to remind 3 of this afternoons meeting to be arranged.Therefore, in 3 the moment of this afternoon, vehicle-mounted end triggers a reminder events, and informs the user through the mode of voice reminder, i.e. content through the playing device on the vehicle-mounted end 26 (for example loudspeaker) broadcast " this afternoon 3 meeting is arranged ".
Can find out that from this embodiment speech input device 20, speech engine 22, memory storage 24 and playing device 26 all are integrated on the car-mounted terminal.
In addition, system also comprises a data transmission interface 28, and system is transferred to the voice data in the memory storage 24 in the external unit (for example outer computer) through this data transmission interface 28, can supply user ID or editor.
The 3rd embodiment of the system of voice reminder
Fig. 6 shows the principle of the 3rd embodiment of the system of voice reminder of the present invention.See also Fig. 6, the system of the voice reminder of present embodiment comprises: speech input device 30, denoising device 32, speech engine 34, memory storage 36, playing device 38.
Annexation between these devices is: speech input device 30 couples denoising device 32, and denoising device 32 couples speech engine 34, and speech engine 34 couples memory storage 36, and memory storage 36 couples playing device 38.
The operation logic of the system of the voice reminder of present embodiment is following.
At vehicle-mounted end, an example of speech input device 30 is the audio input interfaces on the car-mounted terminal.The user is through the input of the audio input interface on vehicle-mounted end voice, and for example, the user says " this afternoon, 3 promptings had meeting ", and mobile unit receives these input voice of user.
Carry out noise reduction process by 32 pairs of phonetic entries that receive of denoising device subsequently.
Speech engine 34 be used in the recognizing voice input content information and be stored in the memory storage 36.
At vehicle-mounted end a speech engine 34 is installed, speech engine 34 receives users' input voice, the content that identifies in the voice to be comprised, with these content stores in the memory storage 36 of vehicle-mounted end.
The speech recognition technology of speech engine 34 is existing technology.For example; Speech engine 34 comprises voice acquisition device, front-end processing module, characteristic extracting module, fundamental tone characteristic extracting module, Tone recognition module, training module, acoustic layer identification module, phonetic syntax Understanding Module, nine parts of language understanding module; This acoustic layer identification module adopts the hidden Markov model of representing with segment length's probability of state; The hidden Markov model that promptly distributes based on the segment length; Be called for short the DDBHMM model, the segment length of this model obeys the distribution with convexity, wherein; Voice units such as phoneme in the state in this model and the voice or syllable are corresponding, and the phonic signal character of these voice unit pronunciations is as the observed quantity of corresponding voice unit; The training method step of this training module is following: to the one or more pronunciation sample files that include some specific word of training module input; The proper vector of the series of frames of every words in this document is unit with the frame; Through the training searching algorithm in the training module; Each state to each speech in every frame voice signal is searched for comparison, obtains the Codebook of Vector Quantization (VQ code book) of phonic signal character vector and the DDBHMM model parameter of this specific word, inputs to the acoustic layer identification module; The audio recognition method step of described Chinese continuous speech recognition system is following: the voice signal through voice acquisition device reception people, carry out front-end processing for the voice signal of importing, and carry out the MFCC phonetic feature (based on the phonetic feature of Mel cepstrum coefficient; Mel-Frequency Cepstral Coefficients) extraction of sequence, this MFCC phonetic feature sequence that obtains is admitted to the acoustic layer identification module, through the searching algorithm of acoustic layer identification module; Produce the recognition result of pinyin lattice form, simultaneously, the fundamental tone eigenvector of voice signal also is extracted out; Send into the Tone recognition module, the Tone recognition module is utilized the breakpoint information of fundamental tone characteristic information and phonetic, obtains the tone information of phonetic and joins in the pinyin lattice; Then; Through the phonetic syntax Understanding Module pinyin lattice is pruned, the Syllable Lattice after simplifying is admitted to the language understanding module, is converted into phonetic figure and speech figure; And in speech figure, search for, get the result that understands to the end.
For example, speech engine 34 can identify the particular content of " this afternoon, 3 promptings had meeting ", knows that this is a reminder events, and the content of prompting is " this afternoon 3 meeting is arranged ".
Playing device 38 is reminded according to the content information that identifies, the voice content of the content of prompting for having stored.
Having identified this at the speech engine 34 of vehicle-mounted end is a reminder events, and is to remind 3 of this afternoons meeting to be arranged.Therefore, in 3 the moment of this afternoon, vehicle-mounted end triggers a reminder events, and informs the user through the mode of voice reminder, i.e. content through the playing device on the vehicle-mounted end 38 (for example loudspeaker) broadcast " this afternoon 3 meeting is arranged ".
Can find out that from this embodiment speech input device 30, denoising device 32, speech engine 34, memory storage 36 and playing device 38 all are integrated on the car-mounted terminal.
The foregoing description provides to those of ordinary skills and realizes or use of the present invention; Those of ordinary skills can be under the situation that does not break away from invention thought of the present invention; The foregoing description is made various modifications or variation; Thereby protection scope of the present invention do not limit by the foregoing description, and should be the maximum magnitude that meets the inventive features that claims mention.
Claims (9)
1. the method for a voice reminder comprises:
The input of reception user's voice;
According to phonetic entry identification wherein content information and store;
Content information according to identifying is reminded, the voice content of the content of prompting for having stored.
2. the method for voice reminder according to claim 1 is characterized in that, receives phonetic entry, identification content information and stores, reminds the voice content of having stored all to accomplish at car-mounted terminal.
3. the method for voice reminder according to claim 2 is characterized in that, the content information that is stored in the phonetic entry of car-mounted terminal exports to computer end.
4. the method for voice reminder according to claim 1 is characterized in that, after receiving the user's voice input and before carrying out content recognition according to phonetic entry, also comprises noise reduction process is carried out in phonetic entry.
5. the method for voice reminder according to claim 2 is characterized in that, the mode of prompting is that the mode that adopts the loudspeaker of car-mounted terminal to carry out voice playing realizes.
6. the system of a voice reminder comprises:
Speech input device receives the voice data that the user imports;
Speech engine couples speech input device, the content information that the recognizing voice input is comprised;
Memory storage couples speech engine, the content information that phonetic entry comprised of storaged voice engine output;
Playing device couples this memory storage, reminds according to the content information that identifies, and the content of prompting is the voice content of having stored in the memory storage.
7. the system of voice reminder according to claim 6 is characterized in that, speech input device, speech engine, memory storage, playing device are integrated in car-mounted terminal.
8. the system of voice reminder according to claim 6 is characterized in that, this system also comprises:
Data transmission interface couples memory storage, and the data in the memory storage is transferred to external unit.
9. the system of voice reminder according to claim 6 is characterized in that, this system also comprises:
Denoising device couples speech input device and speech engine, and noise reduction process is carried out in phonetic entry.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106220609A CN102542705A (en) | 2010-12-31 | 2010-12-31 | Voice reminding method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106220609A CN102542705A (en) | 2010-12-31 | 2010-12-31 | Voice reminding method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102542705A true CN102542705A (en) | 2012-07-04 |
Family
ID=46349504
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010106220609A Pending CN102542705A (en) | 2010-12-31 | 2010-12-31 | Voice reminding method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102542705A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105813005A (en) * | 2014-12-29 | 2016-07-27 | 中国移动通信集团公司 | Method for realizing vehicle-borne reminding, device and terminal |
| CN106204982A (en) * | 2016-08-31 | 2016-12-07 | 惠州学院 | A kind of voice reminder |
| CN106559565A (en) * | 2016-11-04 | 2017-04-05 | 珠海市魅族科技有限公司 | Pronunciation inputting method and electronic equipment |
| CN108810852A (en) * | 2018-06-12 | 2018-11-13 | 奇瑞汽车股份有限公司 | The method and apparatus for creating reminder events |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2691176Y (en) * | 2004-04-08 | 2005-04-06 | 傅家林 | Voice identification vehicle carried telephone |
| CN1941079A (en) * | 2005-09-27 | 2007-04-04 | 通用汽车公司 | Speech recognition method and system |
| CN101001294A (en) * | 2006-12-19 | 2007-07-18 | 中山大学 | Intelligent household voice report and attention system based on voice recognition technology |
| CN101415038A (en) * | 2008-11-21 | 2009-04-22 | 深圳华为通信技术有限公司 | Voice words-leaving method and data card for implementing voice leave word |
| CN201544804U (en) * | 2009-11-19 | 2010-08-11 | 浙江吉利汽车研究院有限公司 | Voice reminding type automobile liquid crystal display system |
-
2010
- 2010-12-31 CN CN2010106220609A patent/CN102542705A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2691176Y (en) * | 2004-04-08 | 2005-04-06 | 傅家林 | Voice identification vehicle carried telephone |
| CN1941079A (en) * | 2005-09-27 | 2007-04-04 | 通用汽车公司 | Speech recognition method and system |
| CN101001294A (en) * | 2006-12-19 | 2007-07-18 | 中山大学 | Intelligent household voice report and attention system based on voice recognition technology |
| CN101415038A (en) * | 2008-11-21 | 2009-04-22 | 深圳华为通信技术有限公司 | Voice words-leaving method and data card for implementing voice leave word |
| CN201544804U (en) * | 2009-11-19 | 2010-08-11 | 浙江吉利汽车研究院有限公司 | Voice reminding type automobile liquid crystal display system |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105813005A (en) * | 2014-12-29 | 2016-07-27 | 中国移动通信集团公司 | Method for realizing vehicle-borne reminding, device and terminal |
| CN106204982A (en) * | 2016-08-31 | 2016-12-07 | 惠州学院 | A kind of voice reminder |
| CN106559565A (en) * | 2016-11-04 | 2017-04-05 | 珠海市魅族科技有限公司 | Pronunciation inputting method and electronic equipment |
| CN108810852A (en) * | 2018-06-12 | 2018-11-13 | 奇瑞汽车股份有限公司 | The method and apparatus for creating reminder events |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN201919034U (en) | Network-based voice prompt system | |
| US8019604B2 (en) | Method and apparatus for uniterm discovery and voice-to-voice search on mobile device | |
| EP2862164B1 (en) | Multiple pass automatic speech recognition | |
| CN111261144B (en) | Voice recognition method, device, terminal and storage medium | |
| EP2252995B1 (en) | Method and apparatus for voice searching for stored content using uniterm discovery | |
| US9484027B2 (en) | Using pitch during speech recognition post-processing to improve recognition accuracy | |
| CN110473546B (en) | Media file recommendation method and device | |
| CN108242236A (en) | Dialog process device and its vehicle and dialog process method | |
| US20180074661A1 (en) | Preferred emoji identification and generation | |
| CN105448294A (en) | Intelligent voice recognition system for vehicle equipment | |
| CN101286317B (en) | Speech recognition device, model training method and traffic information service platform | |
| CN104078044A (en) | Mobile terminal and sound recording search method and device of mobile terminal | |
| CN106710585B (en) | Polyphone broadcasting method and system during interactive voice | |
| US20100178956A1 (en) | Method and apparatus for mobile voice recognition training | |
| CN110097870A (en) | Method of speech processing, device, equipment and storage medium | |
| CN102571882A (en) | Network-based voice reminding method and system | |
| CN112927674B (en) | Voice style migration method and device, readable medium and electronic equipment | |
| EP1374228B1 (en) | Method and processor system for processing of an audio signal | |
| CN112365878A (en) | Speech synthesis method, device, equipment and computer readable storage medium | |
| CN1731511A (en) | Method and system for performing speech recognition on multi-language name | |
| CN101825953A (en) | Chinese character input product with combined voice input and Chinese phonetic alphabet input functions | |
| CN102542705A (en) | Voice reminding method and system | |
| CN114120979A (en) | Optimization method, training method, device and medium of voice recognition model | |
| CN112906369A (en) | Lyric file generation method and device | |
| CN114049875A (en) | TTS (text to speech) broadcasting method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120704 |