CN115171653B - A method, device and computer equipment for reminding pronunciation of uncommon characters - Google Patents
A method, device and computer equipment for reminding pronunciation of uncommon characters Download PDFInfo
- Publication number
- CN115171653B CN115171653B CN202210574787.7A CN202210574787A CN115171653B CN 115171653 B CN115171653 B CN 115171653B CN 202210574787 A CN202210574787 A CN 202210574787A CN 115171653 B CN115171653 B CN 115171653B
- Authority
- CN
- China
- Prior art keywords
- word
- uncommon
- words
- common
- character
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/005—Language recognition
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/04—Segmentation; Word boundary detection
- G10L15/05—Word boundary detection
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/225—Feedback of the input speech
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Machine Translation (AREA)
Abstract
The invention relates to a method and a device for reminding pronunciation of rare words and computer equipment. The method for reminding the pronunciation of the uncommon words comprises the steps of S10, displaying a speakable text, S20, receiving the speakable voice, recognizing the characters corresponding to the speakable voice, tracking the characters in the speakable text word by word, S30, taking the first character to be spoken as the uncommon word according to the tracking progress when judging that the speakable parameters meet the stagnation condition, and reminding the pronunciation of the uncommon word. According to the invention, when the fact that the reader encounters the uncommon word in the reading process is detected, the uncommon word can be automatically sounded and reminded, and the reader does not need to query by oneself, so that the reading consistency and the reading efficiency are improved.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a method and a device for reminding pronunciation of rarely used words and computer equipment.
Background
In a school learning or self-learning process, people often need to read an article or a paragraph. In the reading process, if the uncommon word is encountered, the reading of the uncommon word can be continued by checking a dictionary or surfing the internet to inquire the pronunciation of the uncommon word, and the reading of the speaker is inconvenient.
Disclosure of Invention
The invention aims to solve the technical problem of providing an improved method and device for reminding pronunciation of rarely used words and computer equipment.
The technical scheme adopted by the invention for solving the technical problems is that a method for reminding pronunciation of rarely used words is constructed and comprises the following steps:
s10, displaying a reading text;
s20, receiving a speaking voice, recognizing characters corresponding to the speaking voice, and tracking the characters in the speaking text word by word;
and S30, when the reading parameters are judged to meet the stagnation condition, the first character to be read is used as the uncommon word according to the tracking progress, and pronunciation reminding is carried out on the uncommon word.
Preferably, the pronunciation reminding of the uncommon word comprises pronunciation reminding of the uncommon word with a homophone with the uncommon word, wherein the homophone is more commonly used than the uncommon word.
Preferably, the pronunciation reminding of the uncommon word comprises pronunciation reminding of the uncommon word by pinyin of the uncommon word and/or pronunciation reminding of the uncommon word by voice.
Preferably, the method further comprises the steps of dividing each character into common levels, dividing characters with high common levels into common levels more common than characters with low common levels, grouping homophones in the characters with the divided common levels to form a homophone group library, or
Dividing the words in each group into common levels, wherein the words with high common levels are more common than the words with low common levels, so as to form a homophone group library;
In step S30, the pronunciation reminding is carried out on the uncommon word by homophones with the uncommon word, wherein the homophones are more commonly used relative to the uncommon word, and the method comprises the following steps:
And searching homophones in the homophone group library, which are in the same group as the uncommon words, and acquiring homophones with common level higher than that of the uncommon words to pronounce and remind the uncommon words.
Preferably, each word comprises a sample text which is extracted by self, and the sample text is subjected to single word cutting statistics to obtain each word, or the third party statistical data of an input method is obtained to obtain each word;
when each character is obtained by extracting a sample text by itself, dividing the common level of each character comprises dividing the common level of each character according to the occurrence times of each character;
when each character is obtained by obtaining the third party statistic data of the input method, the division of the common level of each character comprises the division of the common level of each character according to the use times of each character.
Preferably, acquiring homophones with a higher common level than the uncommon words, and performing pronunciation reminding on the uncommon words, including:
And acquiring a plurality of homophones with higher common level than the uncommon words to carry out alternate pronunciation reminding on the uncommon words, wherein the words with higher common level in the homophones are preferentially reminded.
Preferably, the method further comprises the steps of receiving test reading voice, and obtaining a normal time interval and a normal reading speed in a normal reading state according to the test reading voice;
The reading parameter is the current dead time and the dead condition is the normal time interval, and the step S30 includes the steps of when the reading parameter is judged to meet the dead condition, or when the current dead time is judged to be larger than the normal time interval
The step S30 includes when the speakable parameter is judged to meet the stagnation condition, judging that the current speakable speed is smaller than the normal speakable speed.
Preferably, the reminding mode for reminding the uncommon word by pronouncing homonym with the uncommon word comprises the step of displaying the homonym above the uncommon word.
The invention also constructs a rarely used word pronunciation reminding device, which comprises:
The display unit is used for displaying the reading text;
the tracking unit is used for receiving the speaking voice, recognizing the characters corresponding to the speaking voice and tracking the characters in the speaking text word by word;
and the execution unit is used for taking the first character to be read as a rarely used character according to the tracking progress when judging that the reading parameters meet the stagnation condition, and carrying out pronunciation reminding on the rarely used character.
The invention also constructs a computer device comprising a processor and a memory, the processor being communicatively coupled to the memory;
the memory is used for storing a computer program;
the processor is used for executing the computer program stored in the memory to realize the rarely used word pronunciation reminding method.
The implementation of the method, the device and the computer equipment for reminding the pronunciation of the rarely used words has the following beneficial effects:
When detecting that a reader encounters a rare word in the reading process, the method can automatically remind the rare word in a sounding way without the help of a reader to query, so that the consistency and the reading efficiency of reading are improved.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flow chart of the rarely used word pronunciation reminding method of the invention;
fig. 2 is a schematic structural diagram of the rarely used word pronunciation reminding device of the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, fig. 1 is a flow chart of the method for reminding the pronunciation of the uncommon word according to the present invention, as shown in fig. 1, the method for reminding the pronunciation of the uncommon word includes the following steps:
And step S10, displaying the reading text.
It can be understood that the reading text is a text which can be read by people, the reading text can be document books such as articles, news and the like, and the reading text can comprise contents such as text contents, chart contents and the like.
In step S10, a reading text required by the reader to read is displayed for the reader to watch, read according to the reading text, and remind the reader to watch the uncommon words. It will be appreciated that the speakable text may be displayed in accordance with the reader's selection determination.
And step S20, receiving the speakable voice, recognizing the characters corresponding to the speakable voice, and tracking the characters in the speakable text word by word.
It can be understood that the text content in the speakable text is composed of a plurality of words, and when people speak the text content, the words in the text content are sequentially spoken according to the line direction of the text content. For example, from the characteristics of Chinese characters and physiological habits of people, the line direction of the text content is generally from top to bottom and from left to right, and this embodiment will be described by taking this line direction as an example. Of course, the line direction of some text content can also be from top to bottom and from left to right.
When a reader reads the read-aloud text, the words in the word contents are read one by one according to the word direction of the word contents of the read-aloud text, namely the words are read one by one from top to bottom and from left to right. In the process of speaking by a reader, receiving speaking voice, recognizing characters corresponding to the speaking voice, and tracking the characters in the speaking text word by word. Specifically, receiving the speakable voice, recognizing the text corresponding to the speakable voice by adopting a voice-to-text technology, and tracking the text in the speakable text word by matching the recognized text with the text in the text content.
It can be understood that the tracking process can be also understood as gradually adjusting the tracking speed according to the speaking speed of the current reader, that is, according to the condition that the speaking voice is converted into characters, and tracking the characters in the speaking text one by adopting the speed suitable for the speaking speed according to the line direction of the character content, so that the tracked characters are synchronous with the speaking characters. It can be appreciated that the technology of converting voice into text can be directly implemented by adopting the prior art, and the prior technology of converting voice into text can be used for converting third party voice such as hundred degrees, signal flight and the like into text.
Of course, the adjustment of the tracking progress can also be performed according to the adjustment instruction by receiving the adjustment instruction input by the reader. For example, the adjustment instructions may include pause, fast forward, etc., which may be implemented by a key press, etc. In some implementations, the cursor position is consistent with the tracking progress, and thus, the current tracking progress may be adjusted by changing the position of the cursor.
Wherein the current text being tracked may be highlighted in other colors, other fonts, or other representations to represent the tracking or speakable progress.
And step S30, when the reading parameters are judged to meet the stagnation condition, the first character to be read is taken as the uncommon character according to the tracking progress, and the uncommon character is reminded in a pronunciation manner.
It will be appreciated that when the reader's speakable parameters satisfy the stall condition, the reader is considered to be stalled while the tracking is stalled.
The rarely used word pronunciation reminding method further comprises the steps of receiving test reading voice, and obtaining a normal time interval and a normal reading speed in a normal reading state according to the test reading voice.
It can be understood that, before the formal reading, the reader can input a section of voice under the normal reading state as the test reading voice. The test reading voice can be input by normally reading all or part of the content of a reading text by a reader, or by playing the speaking voice of the reader in a normal speaking state, or by directly inputting corresponding voice, the voice input by the reader is used as the test reading voice. It will be appreciated that the normal reading and speaking are described herein to obtain the normal speech rate of the reader, and not the speech rate in the pause state.
And receiving the test reading voice, and acquiring the normal time interval and the normal reading speed of the reader in the normal reading state according to the test reading voice. It can be appreciated that the normal time interval can be obtained by calculating the time interval between two words in the speaking process during normal speaking or normal speaking of the reader. The normal reading speed can be obtained through the nominal word number of the reader for completing the reading or speaking in the nominal time in the normal reading or normal speaking process of the reader. It is understood that the normal time interval, normal read-aloud speed, may be a specific value or interval of values.
In step S30, the speakable parameter may be a current dead time of speaking, and the dead condition may be a normal time interval. It can be understood that when the speakable parameters are judged to meet the stagnation condition, the method includes that when the current stagnation time is judged to be greater than a normal time interval, the speakable is in a stagnation state, the speakable is considered to encounter the uncommon word, the first word in the text to be read behind the tracked current word is taken as the uncommon word according to the text direction, and pronunciation reminding is carried out on the uncommon word.
Or in step S30, the speakable parameter may be the current speakable speed, and the stagnation condition may be the normal speakable speed. It can be understood that when the speakable parameters are judged to meet the stagnation condition, the method comprises the steps of considering that when the current speakable speed is judged to be smaller than the normal speakable speed, namely when the speakable person fails to finish speaking of the rated word number within the rated time, considering that the current speakable speed is smaller than the normal speakable speed, indicating that the speakable person is in a stagnation state, considering that the speakable person encounters the uncommon word, and performing pronunciation reminding on the uncommon word.
In the embodiment, the method for reminding the pronunciation of the uncommon word comprises the step of reminding the pronunciation of the uncommon word by the homophone word homophone with the uncommon word, wherein the homophone word is more commonly used than the uncommon word.
The method for reminding the pronunciation of the uncommon words further comprises the steps of S01, dividing each word into common levels, wherein the words with high common levels are more common than the words with low common levels, and grouping homophones in the words with the divided common levels to form a homophone group library.
This step S01 may precede step S10. It can be appreciated that each word is processed before the reader reads, so that homophones can be used to sound the rarely used words.
The method comprises the steps of obtaining each character, namely, obtaining each character by automatically extracting a sample text and performing single word cutting statistics on the character content in the sample text, or obtaining the third party statistical data of an input method.
When each character is obtained by extracting the sample text by self, the division of the common level of each character comprises the division of the common level of each character according to the occurrence times of each character. It can be understood that the sample text can be a document reading such as news, articles and the like, and one or more sample texts can be adopted, and the text content in the sample text contains more text as much as possible, so that the text is more comprehensive.
It can be understood that the word cutting statistics, for example, the word content includes words such as "newspaper, storm, leopard, etc., can be cut into words such as" newspaper "," holding "," storm "," leopard, ancient "and count the number of times of occurrence of each word, for example, if the words such as" newspaper "," holding "," storm "," leopard, ancient "occur 30 times, 20 times, 15 times, 8 times, 10 times respectively, the common level can be used to distinguish the common level of each word, where the common level of the word with higher occurrence number is higher than the common level of the word with lower occurrence number. The usage level may include a plurality of, for example, the usage level includes a usage level 1 to a usage level 10 total of 10 usage levels, with smaller numbers of usage levels representing higher usage levels.
When each character is obtained by obtaining the third party statistic data of the input method, the division of the common level of each character comprises the division of the common level of each character according to the use times of each character. The input method can be a dog searching and hundred-degree input method, and the common level with higher use times is higher than the common level of characters with lower use times.
After the division of the common level of each word is completed, homophones in the words with the divided common level are grouped to form a homophone group library. It is understood that homophones are pinyin identical and intonation identical. For example, the pronunciation of "newspaper", "holding", "riot" and "leopard" in the text content is "b-a" so that the plurality of characters are homophones, and the plurality of homophones are divided into a homophone group, so that the homophone group includes homophones "newspaper", "holding", "riot" and "leopard" in the common level 2, the common level 4, the common level 6 and the common level 8 respectively, and the homophones can be ordered according to the common level. It will be appreciated that the common level of homophones in the homophones set is for illustration only and does not represent an actual common level. The plurality of homophones are so arranged as to form a homophone library.
In some embodiments, step S01 may be to group homophones in each word, and divide each word in each group into common levels, where words with high common levels are more common than words with low common levels, so as to form a homophone group library.
It will be understood that the step S01 in this embodiment is different from the step S01 in this embodiment in that in this embodiment, the common level division of each word is completed first, and then homophones in the words with the divided common levels are grouped. In some embodiments, homophones in each word are grouped, and then the words in each group are divided into common levels. It will be appreciated that the two are merely in a different order of execution. In the invention, the execution sequence of the two steps is not limited, as long as each homophone group can be formed in the homophone group library, and each word in each homophone group has a corresponding common level, and of course, the words in each homophone group can be ordered according to the common level.
In step S30, pronunciation reminding is carried out on the uncommon words by homophones which are homophones with the uncommon words, wherein homophones are more commonly used than the uncommon words. Specifically, when the rarely used word is identified, the same homophone word as the rarely used word is searched in the homophone group library, and homophone words with higher common level than the rarely used word are obtained to sound and remind the rarely used word.
Further, acquiring homophones with common levels higher than those of the uncommon words to sound the uncommon words, and acquiring a plurality of homophones with common levels higher than those of the uncommon words to sound the uncommon words in a replacement way, wherein the characters with the common levels higher than those of the uncommon words are preferentially reminded.
Specifically, all homophones with higher common level than the uncommon words are searched in the homophone group library, and several homophones with higher common level are obtained, for example, three homophones with highest common level are obtained, and alternate pronunciation reminding is carried out on the uncommon words.
Of course, other numbers of homophones may be used for replacement reminding, for example, when the common level is more than three homophones higher than the uncommon word, the homophones are replaced at intervals, or the first homophones with higher common levels are replaced at intervals, and when the common level is less than three homophones with higher common levels than the uncommon word, all homophones are replaced at intervals.
The interval replacement pronunciation reminding mode comprises three homophone word timing interval replacement, for example, when the identified rare word is leopard, the homophone group with the word leopard is searched in the homophone group library, and homophone group newspaper, embracing, riot and leopard with the word leopard are searched. In the homophone group, the words of report, embrace and riot are higher than the common level of the words of leopard, so that pronunciation reminding can be carried out by alternately replacing the words of report, embrace and riot.
For example, the homonym "newspaper" with the highest common level is used for pronunciation reminding, the homonym "holding" with the common level is used for pronunciation reminding after the set timing time is reached, and the homonym "riot" with the common level is used for pronunciation reminding after the set timing time is reached after the homonym "holding" is started. And after the set timing time is reached from the start of 'riot' reminding, if the reader is judged to still not resume reading, the 'newspaper' is used for sounding reminding again, and reminding is circulated until the reader resumes reading. It will be appreciated that the timing time may be set as desired.
The timing time may be a time for judging whether the speaker is still in a stagnation state, and it can be understood that in step S30, a homophone is adopted to sound the uncommon word to remind to start timing, if the speaker is still in a stagnation state after the timing time is reached, the homophone is considered to still belong to the uncommon word for the speaker, so that another homophone is replaced to sound the uncommon word. The timing time may be the same as the normal time interval or greater than the normal time interval to provide the reader with more discrimination time. Or the timing time is less than the normal time interval, e.g., the timing time is one third of the normal time interval, to display more homophones to the speakerphone.
Of course, the determination of whether the reader is in a stagnation state may be implemented by the method of determining whether the current reading speed is smaller than the normal reading speed, which is the same as the above determination, and no repeated description is given here.
In some embodiments, the alternate pronunciation reminding mode includes that the replacement reminding of homophones is carried out by receiving the replacement instruction input by the reader, and similarly, the replacement of homophones is carried out by preferentially displaying the frequently used level to be high, namely, reminding in the sequence from the frequently used level to be low.
According to the application, the homophones with high common level, namely homophones with small common level numbers, are used for carrying out pronunciation reminding on the uncommon words preferentially, and the homophones are used for carrying out pronunciation reminding on the uncommon words, so that the reading efficiency of a reader can be improved. In other embodiments, multiple homophones may be used simultaneously to sound the alert of the uncommon word.
The homophone is used for reminding the uncommon words by pronunciation, and the homophone is displayed above the uncommon words. It can be understood that in the text content, a space is reserved between each row of text, homophones are displayed above the uncommon words, so that the words are convenient for a reader to check, and the words accord with daily habits.
It will be appreciated that the homophone gallery may be entered into the speakable device prior to the speakable text being spoken by the reader, so that the pronunciation of the uncommon word may be alerted by the homophone gallery during the speakable process of the reader, and the homophone word may be presented in a highlighted presentation, e.g., the homophone word may be different from other words in the speakable text in color, thickness, font, etc.
It will be appreciated that the same reader will have different speaking speeds for different speakable texts and that different speakers will have different speaking speeds for the same speakable text. Therefore, in step S20, the text in the read-aloud text is tracked according to the read-aloud speed of the current reader, namely the condition that the voice is converted into the text, and in step S30, the stagnation condition is set according to the read-aloud speed of the current reader.
In the embodiment, based on the reading speed of the reader, the pronunciation of the rarely used words is automatically prompted, homophones can be dynamically replaced according to the stagnation condition of the reader, and the reading consistency and the reading efficiency of the reader can be improved.
In another embodiment of the present invention, the method for alert pronunciation of rarely used words comprises the steps of:
And step S10, displaying the reading text.
And step S20, receiving the speakable voice, recognizing the characters corresponding to the speakable voice, and tracking the characters in the speakable text word by word.
And step S30, when the reading parameters are judged to meet the stagnation condition, the first character to be read is taken as the uncommon character according to the tracking progress, and the uncommon character is reminded in a pronunciation manner.
The difference between this embodiment and the above embodiment is that in step S30, the pronunciation reminding of the uncommon word includes pronunciation reminding of the uncommon word with homophone with the uncommon word, pronunciation reminding of the uncommon word with pinyin of the uncommon word, and/or pronunciation reminding of the uncommon word with voice.
It will be appreciated that the method of the above embodiment is the same as that of the previous embodiment in that the pronunciation of the uncommon word is reminded by homophones that homophones the uncommon word. The pronunciation reminding method comprises the steps of reminding the rare word by using the pinyin of the rare word, specifically, establishing a Chinese pinyin library of each character in advance, searching the Chinese pinyin corresponding to the rare word from the Chinese pinyin library when the rare word is identified, and displaying the Chinese pinyin corresponding to the rare word above the rare word. Of course, the Chinese pinyin library can be implemented by using the existing Chinese pinyin library.
Pronunciation reminding is carried out on the rarely used words in a voice mode, specifically, the sounding unit is arranged, and when the rarely used words are identified, the rarely used words are broadcasted through the sounding unit.
It can be appreciated that the above-mentioned one or more combination modes can be adopted to sound the remote words for reminding, so as to be suitable for various crowds.
Referring to fig. 2, fig. 2 is a schematic structural diagram of the rarely used word pronunciation reminding device according to the present invention, and the rarely used word pronunciation reminding device 10 performs functions corresponding to the rarely used word pronunciation reminding method. As shown in fig. 2, the rarely used word pronunciation reminding apparatus 10 includes:
And a display unit 101 for displaying the speakable text.
It can be understood that the reading text is a text which can be read by people, the reading text can be document books such as articles, news and the like, and the reading text can comprise contents such as text contents, chart contents and the like.
In the display unit 101, a speakable text that the reader needs to speak is displayed for the reader to view, and speaks according to the speakable text, and a reminder for the reader to view the uncommon word. It will be appreciated that the speakable text may be displayed in accordance with the reader's selection determination.
And the tracking unit 102 is used for receiving the speakable voice, recognizing the characters corresponding to the speakable voice and tracking the characters in the speakable text word by word.
It can be understood that the text content in the speakable text is composed of a plurality of words, and when people speak the text content, the words in the text content are sequentially spoken according to the line direction of the text content. For example, from the characteristics of Chinese characters and physiological habits of people, the line direction of the text content is generally from top to bottom and from left to right, and this embodiment will be described by taking this line direction as an example. Of course, the line direction of some text content can also be from top to bottom and from left to right.
When a reader reads the read-aloud text, the words in the word contents are read one by one according to the word direction of the word contents of the read-aloud text, namely the words are read one by one from top to bottom and from left to right. In the process of speaking by a reader, receiving speaking voice, recognizing characters corresponding to the speaking voice, and tracking the characters in the speaking text word by word. Specifically, receiving the speakable voice, recognizing the text corresponding to the speakable voice by adopting a voice-to-text technology, and tracking the text in the speakable text word by matching the recognized text with the text in the text content.
It can be understood that the tracking process can be also understood as gradually adjusting the tracking speed according to the speaking speed of the current reader, that is, according to the condition that the speaking voice is converted into characters, and tracking the characters in the speaking text one by adopting the speed suitable for the speaking speed according to the line direction of the character content, so that the tracked characters are synchronous with the speaking characters. It can be appreciated that the technology of converting voice into text can be directly implemented by adopting the prior art, and the prior technology of converting voice into text can be used for converting third party voice such as hundred degrees, signal flight and the like into text.
Of course, the adjustment of the tracking progress can also be performed according to the adjustment instruction by receiving the adjustment instruction input by the reader. For example, the adjustment instructions may include pause, fast forward, etc., which may be implemented by a key press, etc. In some implementations, the cursor position is consistent with the tracking progress, and thus, the current tracking progress may be adjusted by changing the position of the cursor.
Wherein the current text being tracked may be highlighted in other colors, other fonts, or other representations to represent the tracking or speakable progress.
And the execution unit 103 is used for taking the first character to be read as the uncommon word according to the tracking progress and carrying out pronunciation reminding on the uncommon word when judging that the reading parameter meets the stagnation condition.
It will be appreciated that when the reader's speakable parameters satisfy the stall condition, the reader is considered to be stalled while the tracking is stalled.
The rarely used word pronunciation reminding device 10 of the embodiment further includes a receiving and obtaining unit, configured to receive the test reading voice, and obtain a normal time interval and a normal reading speed in a normal reading state according to the test reading voice.
It can be understood that, before the formal reading, the reader can input a section of voice under the normal reading state as the test reading voice. The test reading voice can be input by normally reading all or part of the content of a reading text by a reader, or by playing the speaking voice of the reader in a normal speaking state, or by directly inputting corresponding voice, the voice input by the reader is used as the test reading voice. It will be appreciated that the normal reading and speaking are described herein to obtain the normal speech rate of the reader, and not the speech rate in the pause state.
And receiving the test reading voice, and acquiring the normal time interval and the normal reading speed of the reader in the normal reading state according to the test reading voice. It can be appreciated that the normal time interval can be obtained by calculating the time interval between two words in the speaking process during normal speaking or normal speaking of the reader. The normal reading speed can be obtained through the nominal word number of the reader for completing the reading or speaking in the nominal time in the normal reading or normal speaking process of the reader. It is understood that the normal time interval, normal read-aloud speed, may be a specific value or interval of values.
In the execution unit 103, the speakable parameter may be a current dead time of the speakable, and the dead condition may be a normal time interval. It can be understood that when the speakable parameters are judged to meet the stagnation condition, the method includes that when the current stagnation time is judged to be greater than a normal time interval, the speakable is in a stagnation state, the speakable is considered to encounter the uncommon word, the first word in the text to be read behind the tracked current word is taken as the uncommon word according to the text direction, and pronunciation reminding is carried out on the uncommon word.
Or in the execution unit 103, the speakable parameter may be a current speakable speed, and the stagnation condition may be a normal speakable speed. It can be understood that when the speakable parameters are judged to meet the stagnation condition, the method comprises the steps of considering that when the current speakable speed is judged to be smaller than the normal speakable speed, namely when the speakable person fails to finish speaking of the rated word number within the rated time, considering that the current speakable speed is smaller than the normal speakable speed, indicating that the speakable person is in a stagnation state, considering that the speakable person encounters the uncommon word, and performing pronunciation reminding on the uncommon word.
In the embodiment, the method for reminding the pronunciation of the uncommon word comprises the step of reminding the pronunciation of the uncommon word by the homophone word homophone with the uncommon word, wherein the homophone word is more commonly used than the uncommon word.
The rarely used word pronunciation reminding device 10 of the embodiment further comprises a grouping and dividing unit, wherein the grouping and dividing unit is used for dividing each word into common levels, words with high common levels are more common than words with low common levels, and homophones in the words with the divided common levels are grouped to form a homophone group library.
The grouping dividing unit may be performed before the display unit 101. It can be appreciated that each word is processed before the reader reads, so that homophones can be used to sound the rarely used words.
The method comprises the steps of obtaining each character by automatically extracting a sample text and performing single word cutting statistics on the character content in the sample text, or obtaining the third party statistical data of an input method to obtain each character.
When each character is obtained by extracting the sample text by self, the division of the common level of each character comprises the division of the common level of each character according to the occurrence times of each character. It can be understood that the sample text can be a document reading such as news, articles and the like, and one or more sample texts can be adopted, and the text content in the sample text contains more text as much as possible, so that the text is more comprehensive.
It can be understood that the word cutting statistics, for example, the word content includes words such as "newspaper, storm, leopard, etc., can be cut into words such as" newspaper "," holding "," storm "," leopard, ancient "and count the number of times of occurrence of each word, for example, if the words such as" newspaper "," holding "," storm "," leopard, ancient "occur 30 times, 20 times, 15 times, 8 times, 10 times respectively, the common level can be used to distinguish the common level of each word, where the common level of the word with higher occurrence number is higher than the common level of the word with lower occurrence number. The usage level may include a plurality of, for example, the usage level includes a usage level 1 to a usage level 10 total of 10 usage levels, with smaller numbers of usage levels representing higher usage levels.
When each character is obtained by obtaining the third party statistic data of the input method, the division of the common level of each character comprises the division of the common level of each character according to the use times of each character. The input method can be a dog searching and hundred-degree input method, and the common level with higher use times is higher than the common level of characters with lower use times.
After the division of the common level of each word is completed, homophones in the words with the divided common level are grouped to form a homophone group library. It is understood that homophones are pinyin identical and intonation identical. For example, the pronunciation of "newspaper", "holding", "riot" and "leopard" in the text content is "b-a" so that the plurality of characters are homophones, and the plurality of homophones are divided into a homophone group, so that the homophone group includes homophones "newspaper", "holding", "riot" and "leopard" in the common level 2, the common level 4, the common level 6 and the common level 8 respectively, and the homophones can be ordered according to the common level. It will be appreciated that the common level of homophones in the homophones set is for illustration only and does not represent an actual common level. The plurality of homophones are so arranged as to form a homophone library.
In some embodiments, the grouping unit groups homophones in each word, and the common level of each word in each group is divided, and words with high common level are more common than words with low common level, so as to form a homophone group library.
It will be understood that the grouping unit in this embodiment is different from the grouping unit in this embodiment in that in this embodiment, the common level division of each word is completed first, and then homophones in the words of which the common level has been divided are grouped. In some embodiments, homophones in each word are grouped, and then the words in each group are divided into common levels. It will be appreciated that the two are merely in a different order of execution. In the invention, the execution sequence of the two steps is not limited, as long as each homophone group can be formed in the homophone group library, and each word in each homophone group has a corresponding common level, and of course, the words in each homophone group can be ordered according to the common level.
In the execution unit 103, the pronunciation reminding is performed on the uncommon word with the homophone of the uncommon word, wherein the homophone is more commonly used than the uncommon word. Specifically, when the rarely used word is identified, the same homophone word as the rarely used word is searched in the homophone group library, and homophone words with higher common level than the rarely used word are obtained to sound and remind the rarely used word.
Further, acquiring homophones with common levels higher than those of the uncommon words to sound the uncommon words, and acquiring a plurality of homophones with common levels higher than those of the uncommon words to sound the uncommon words in a replacement way, wherein the characters with the common levels higher than those of the uncommon words are preferentially reminded.
Specifically, all homophones with higher common level than the uncommon words are searched in the homophone group library, and several homophones with higher common level are obtained, for example, three homophones with highest common level are obtained, and alternate pronunciation reminding is carried out on the uncommon words.
Of course, other numbers of homophones may be used for replacement reminding, for example, when the common level is more than three homophones higher than the uncommon word, the homophones are replaced at intervals, or the first homophones with higher common levels are replaced at intervals, and when the common level is less than three homophones with higher common levels than the uncommon word, all homophones are replaced at intervals.
The interval replacement pronunciation reminding mode comprises three homophone word timing interval replacement, for example, when the identified rare word is leopard, the homophone group with the word leopard is searched in the homophone group library, and homophone group newspaper, embracing, riot and leopard with the word leopard are searched. In the homophone group, the words of report, embrace and riot are higher than the common level of the words of leopard, so that pronunciation reminding can be carried out by alternately replacing the words of report, embrace and riot.
For example, the homonym "newspaper" with the highest common level is used for pronunciation reminding, the homonym "holding" with the common level is used for pronunciation reminding after the set timing time is reached, and the homonym "riot" with the common level is used for pronunciation reminding after the set timing time is reached after the homonym "holding" is started. And after the set timing time is reached from the start of 'riot' reminding, if the reader is judged to still not resume reading, the 'newspaper' is used for sounding reminding again, and reminding is circulated until the reader resumes reading. It will be appreciated that the timing time may be set as desired.
The timing time may be a time for judging whether the speaker is still in a stagnation state, and it can be understood that, in the execution unit 103, a homophone is adopted to sound the uncommon word to remind to start timing, if the speaker is still in a stagnation state after the timing time is reached, the homophone is considered to still belong to the uncommon word for the speaker, so that another homophone is replaced to sound the uncommon word. The timing time may be the same as the normal time interval or greater than the normal time interval to provide the reader with more discrimination time. Or the timing time is less than the normal time interval, e.g., the timing time is one third of the normal time interval, to display more homophones to the speakerphone.
Of course, the determination of whether the reader is in a stagnation state may be implemented by the method of determining whether the current reading speed is smaller than the normal reading speed, which is the same as the above determination, and no repeated description is given here.
In some embodiments, the alternate pronunciation reminding mode includes that the replacement reminding of homophones is carried out by receiving the replacement instruction input by the reader, and similarly, the replacement of homophones is carried out by preferentially displaying the frequently used level to be high, namely, reminding in the sequence from the frequently used level to be low.
According to the application, the homophones with high common level, namely homophones with small common level numbers, are used for carrying out pronunciation reminding on the uncommon words preferentially, and the homophones are used for carrying out pronunciation reminding on the uncommon words, so that the reading efficiency of a reader can be improved. In other embodiments, multiple homophones may be used simultaneously to sound the alert of the uncommon word.
The homophone is used for reminding the uncommon words by pronunciation, and the homophone is displayed above the uncommon words. It can be understood that in the text content, a space is reserved between each row of text, homophones are displayed above the uncommon words, so that the words are convenient for a reader to check, and the words accord with daily habits.
It will be appreciated that the homophone gallery may be entered into the speakable device prior to the speakable text being spoken by the reader, so that the pronunciation of the uncommon word may be alerted by the homophone gallery during the speakable process of the reader, and the homophone word may be presented in a highlighted presentation, e.g., the homophone word may be different from other words in the speakable text in color, thickness, font, etc.
It will be appreciated that the same reader will have different speaking speeds for different speakable texts and that different speakers will have different speaking speeds for the same speakable text. Therefore, in the tracking unit 102, the text in the speakable text is tracked according to the speaking speed of the current reader, that is, the condition that the voice is converted into the text, and in the executing unit 103, the stagnation condition is set according to the speaking speed of the current reader.
In the embodiment, based on the reading speed of the reader, the pronunciation of the rarely used words is automatically prompted, homophones can be dynamically replaced according to the stagnation condition of the reader, and the reading consistency and the reading efficiency of the reader can be improved.
In another embodiment of the present invention, the rarely used word pronunciation reminding device comprises:
And the display unit is used for displaying the reading text.
And the tracking unit is used for receiving the speakable voice, recognizing the characters corresponding to the speakable voice and tracking the characters in the speakable text word by word.
And the execution unit is used for taking the first character to be read as a rarely used character according to the tracking progress when judging that the reading parameter meets the stagnation condition and carrying out pronunciation reminding on the rarely used character.
The embodiment is different from the embodiment in that the execution unit performs pronunciation reminding on the uncommon word, including pronunciation reminding on the uncommon word with homonym homonymous with the uncommon word, and/or pronunciation reminding on the uncommon word with Pinyin of the uncommon word, and/or pronunciation reminding on the uncommon word with voice.
It will be appreciated that the method for performing pronunciation reminding on the uncommon word by homophones with the uncommon word is the same as the method for performing the foregoing embodiment, and will not be described in detail herein. The pronunciation reminding method comprises the steps of reminding the rare word by using the pinyin of the rare word, specifically, establishing a Chinese pinyin library of each character in advance, searching the Chinese pinyin corresponding to the rare word from the Chinese pinyin library when the rare word is identified, and displaying the Chinese pinyin corresponding to the rare word above the rare word. Of course, the Chinese pinyin library can be implemented by using the existing Chinese pinyin library.
Pronunciation reminding is carried out on the rarely used words in a voice mode, specifically, the sounding unit is arranged, and when the rarely used words are identified, the rarely used words are broadcasted through the sounding unit.
It can be appreciated that the above-mentioned one or more combination modes can be adopted to sound the remote words for reminding, so as to be suitable for various crowds.
The invention provides a computer device which comprises a processor and a memory, wherein the processor is in communication connection with the memory, the memory is used for storing a computer program, and the processor is used for executing the computer program stored in the memory to realize the steps of the rarely used word pronunciation reminding method.
The invention relates to a readable storage medium, which stores a computer program, wherein the computer program realizes the steps of the rarely used word pronunciation reminding method when being executed by a processor.
Those skilled in the art will appreciate that all or part of the procedures in the methods of the above embodiments may be implemented by a computer program for instructing relevant hardware, where the above program may be stored in a readable storage medium of a terminal device such as a computer, a mobile phone, etc., and the program may include the procedures of the embodiments of the above methods when executed. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random-access Memory (Random Access Memory, RAM), or the like.
It is understood that the foregoing examples merely illustrate preferred embodiments of the present invention, and the description thereof is specific and detailed and not to be construed as limiting the invention, and that it is understood that various changes and modifications can be made by one skilled in the art without departing from the spirit of the invention, and that it is intended to cover all modifications and adaptations of the invention as fall within the scope of the invention.
Claims (8)
1. The pronunciation reminding method for the rarely used words is characterized by comprising the following steps of:
S01, dividing each character into common levels, wherein the characters with high common levels are more common than the characters with low common levels, grouping homophones in the characters with the divided common levels to form a homophone group library, or
Dividing the words in each group into common levels, wherein the words with high common levels are more common than the words with low common levels, so as to form a homophone group library;
s10, displaying a reading text;
s20, receiving a speaking voice, recognizing characters corresponding to the speaking voice, and tracking the characters in the speaking text word by word;
And S30, when the reading parameters are judged to meet the stagnation condition, the first character to be read is taken as the uncommon character according to the tracking progress, homophones which are in the same group with the uncommon character are searched in the homophone group library, and homophones with the common level higher than the uncommon character are obtained to carry out pronunciation reminding on the uncommon character.
2. The method of claim 1, wherein the alerting the uncommon word to pronounce comprises alerting the uncommon word to pronounce with pinyin of the uncommon word and/or alerting the uncommon word to pronounce with speech.
3. The rarely used word pronunciation reminding method according to claim 1 is characterized in that each word comprises the steps of automatically extracting a sample text, and performing single word cutting statistics on the sample text to obtain each word, or obtaining third party statistical data of an input method to obtain each word;
when each character is obtained by extracting a sample text by itself, dividing the common level of each character comprises dividing the common level of each character according to the occurrence times of each character;
when each character is obtained by obtaining the third party statistic data of the input method, the division of the common level of each character comprises the division of the common level of each character according to the use times of each character.
4. The method of claim 1, wherein obtaining homophones with higher common level than the uncommon words to pronounce the uncommon words comprises:
And acquiring a plurality of homophones with higher common level than the uncommon words to carry out alternate pronunciation reminding on the uncommon words, wherein the words with higher common level in the homophones are preferentially reminded.
5. The method for reminding the pronunciation of the rarely used words according to claim 1 is characterized by further comprising the steps of receiving test reading voice, and obtaining a normal time interval and a normal reading speed in a normal reading state according to the test reading voice;
The reading parameter is the current dead time and the dead condition is the normal time interval, and the step S30 includes the steps of when the reading parameter is judged to meet the dead condition, or when the current dead time is judged to be larger than the normal time interval
The step S30 includes when the speakable parameter is judged to meet the stagnation condition, judging that the current speakable speed is smaller than the normal speakable speed.
6. The method for reminding the pronunciation of the uncommon word according to claim 1, wherein the reminding mode of reminding the pronunciation of the uncommon word by homophones with the uncommon word comprises displaying the homophones above the uncommon word.
7. The utility model provides a rare word pronunciation reminding device which characterized in that includes:
The grouping and dividing unit is used for dividing each character into common levels, wherein the characters with high common levels are more common than the characters with low common levels, grouping homophones in the characters with the divided common levels to form a homophone group library, or
Dividing the words in each group into common levels, wherein the words with high common levels are more common than the words with low common levels, so as to form a homophone group library;
The display unit is used for displaying the reading text;
the tracking unit is used for receiving the speaking voice, recognizing the characters corresponding to the speaking voice and tracking the characters in the speaking text word by word;
and the execution unit is used for taking the first character to be read as a uncommon word according to the tracking progress when judging that the reading parameter meets the stagnation condition, searching homophones of the same group as the uncommon word in the homophone group library, and acquiring homophones with common level higher than the uncommon word to sound and remind the uncommon word.
8. A computer device comprising a processor and a memory, the processor being communicatively coupled to the memory;
the memory is used for storing a computer program;
The processor is configured to execute a computer program stored in the memory to implement the uncommon word pronunciation alert method as set forth in any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210574787.7A CN115171653B (en) | 2022-05-25 | 2022-05-25 | A method, device and computer equipment for reminding pronunciation of uncommon characters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210574787.7A CN115171653B (en) | 2022-05-25 | 2022-05-25 | A method, device and computer equipment for reminding pronunciation of uncommon characters |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115171653A CN115171653A (en) | 2022-10-11 |
| CN115171653B true CN115171653B (en) | 2025-05-23 |
Family
ID=83483443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210574787.7A Active CN115171653B (en) | 2022-05-25 | 2022-05-25 | A method, device and computer equipment for reminding pronunciation of uncommon characters |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115171653B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103971546A (en) * | 2014-04-16 | 2014-08-06 | 无敌科技(西安)有限公司 | Character identification method and device |
| CN113268981A (en) * | 2021-05-27 | 2021-08-17 | 咪咕音乐有限公司 | Information processing method and device and electronic equipment |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE60119643T2 (en) * | 2000-09-18 | 2007-02-01 | L & H Holdings USA, Inc., Burlington | Homophone choice in speech recognition |
| US20120089400A1 (en) * | 2010-10-06 | 2012-04-12 | Caroline Gilles Henton | Systems and methods for using homophone lexicons in english text-to-speech |
| CN103186581A (en) * | 2011-12-30 | 2013-07-03 | 牟颖 | A method for quickly acquiring the pronunciation of rare words in books through mobile phones |
| CN109840294B (en) * | 2018-12-28 | 2023-04-18 | 深圳市世强元件网络有限公司 | Method for inquiring matching data of electronic element, storage medium and terminal |
| CN110853437A (en) * | 2019-11-27 | 2020-02-28 | 墨子(深圳)人工智能技术有限公司 | Intelligent device based on operation tutoring and correcting |
| CN112765445A (en) * | 2021-01-26 | 2021-05-07 | 维沃移动通信有限公司 | Rarely-used word recognition method and device |
| CN113610681A (en) * | 2021-08-17 | 2021-11-05 | 山西传世科技有限公司 | AI-based user interactive reading support method and system |
-
2022
- 2022-05-25 CN CN202210574787.7A patent/CN115171653B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103971546A (en) * | 2014-04-16 | 2014-08-06 | 无敌科技(西安)有限公司 | Character identification method and device |
| CN113268981A (en) * | 2021-05-27 | 2021-08-17 | 咪咕音乐有限公司 | Information processing method and device and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115171653A (en) | 2022-10-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11404043B2 (en) | Systems and methods for providing non-lexical cues in synthesized speech | |
| CN110706536B (en) | Method and device for answering questions by voice | |
| US11935523B2 (en) | Detection of correctness of pronunciation | |
| US9009051B2 (en) | Apparatus, method, and program for reading aloud documents based upon a calculated word presentation order | |
| CN108874935B (en) | A method and electronic device for recommending review content based on voice search | |
| CN104157286B (en) | A kind of phrasal acquisition methods and device | |
| US20180011687A1 (en) | Head-mounted display system and operating method for head-mounted display device | |
| US7406408B1 (en) | Method of recognizing phones in speech of any language | |
| WO2004072780A2 (en) | Method for automatic and semi-automatic classification and clustering of non-deterministic texts | |
| CN111710328B (en) | Training sample selection method, device and medium for speech recognition model | |
| CN110111778B (en) | Voice processing method and device, storage medium and electronic equipment | |
| CN111739556A (en) | System and method for voice analysis | |
| CN104134439B (en) | A kind of phrasal acquisition methods, apparatus and system | |
| CN109948124A (en) | Voice document cutting method, device and computer equipment | |
| US9805740B2 (en) | Language analysis based on word-selection, and language analysis apparatus | |
| KR20190083438A (en) | Korean dialogue apparatus | |
| CN116881412A (en) | Chinese character multidimensional information matching training method and device, electronic equipment and storage medium | |
| CN108959163B (en) | Subtitle display method for audio electronic book, electronic device and computer storage medium | |
| CN115171653B (en) | A method, device and computer equipment for reminding pronunciation of uncommon characters | |
| CN107767862B (en) | Voice data processing method, system and storage medium | |
| US7430503B1 (en) | Method of combining corpora to achieve consistency in phonetic labeling | |
| CN110428668B (en) | Data extraction method and device, computer system and readable storage medium | |
| Hanique et al. | Choice and pronunciation of words: Individual differences within a homogeneous group of speakers | |
| Hlaing et al. | Myanmar speech synthesis system by using phoneme concatenation method | |
| CN114420086B (en) | Speech synthesis method and device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |