WO2018014570A1 - 语音转化手写笔迹的方法和装置 - Google Patents
语音转化手写笔迹的方法和装置 Download PDFInfo
- Publication number
- WO2018014570A1 WO2018014570A1 PCT/CN2017/077049 CN2017077049W WO2018014570A1 WO 2018014570 A1 WO2018014570 A1 WO 2018014570A1 CN 2017077049 W CN2017077049 W CN 2017077049W WO 2018014570 A1 WO2018014570 A1 WO 2018014570A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- handwriting
- data
- handwritten
- voice
- inner code
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012545 processing Methods 0.000 claims abstract description 31
- 238000006243 chemical reaction Methods 0.000 claims description 24
- 239000013589 supplement Substances 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 15
- 230000009469 supplementation Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229920001690 polydopamine Polymers 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Images
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
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/06—Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids
-
- 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
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/06—Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids
- G10L21/10—Transforming into visible information
Definitions
- the present invention relates to the field of information technology, and in particular, to a method and apparatus for voice conversion handwriting.
- an embodiment of the technical solution of the present invention is:
- a method for speech conversion handwriting comprising the steps of:
- the handwriting restoration process is performed on the key data to obtain a handwritten restoration handwriting.
- an apparatus for voice conversion handwriting comprising:
- a voice recognition unit configured to perform voice recognition processing on the voice data when the voice data input by the user is received, to obtain a voice inner code corresponding to the voice data
- a retrieval unit configured to retrieve the inner code of the voice in the handwriting database, and obtain the key data of the handwritten handwriting corresponding to the inner code of the handwriting from the handwriting database when the inner code of the handwriting corresponding to the inner code of the voice is retrieved;
- the handwriting restoration unit is used for handwriting restoration processing on key data to obtain handwritten restoration handwriting.
- the method and device for translating handwritten handwriting of the present invention retrieves the inner code obtained by the voice recognition in the handwriting database, and obtains the key data of the handwriting corresponding to the voice data, thereby associating the voice data with the handwritten handwriting of the user through the inner code,
- the vocabulary content in the voice can be converted into the user's handwriting handwriting; when the user writes on the writable device, the voice data can be converted into a writing handwriting for supplementation by voice input, and the process is simple and convenient to use.
- Embodiment 1 is a schematic flow chart of Embodiment 1 of a method for converting a handwriting handwriting according to the present invention
- Embodiment 2 is a schematic flow chart of Embodiment 2 of a method for converting a handwritten handwriting according to the present invention
- FIG. 3 is a schematic structural diagram of Embodiment 1 of a device for converting a handwritten handwriting according to the present invention.
- FIG. 1 is a schematic flowchart of the method for the voice conversion handwriting of the present invention.
- the method embodiment 1 for voice conversion handwriting can include the following steps:
- Step S110 Perform voice recognition processing on the voice data when the voice data input by the user is received, and obtain a voice inner code corresponding to the voice data;
- Step S120 Retrieving the inner code of the voice in the handwriting database, and acquiring the key data of the handwritten handwriting corresponding to the inner code of the handwriting from the handwriting database when the inner code of the handwriting corresponding to the inner code of the voice is retrieved;
- Step S130 performing handwriting restoration processing on the key data to obtain a handwritten restoration handwriting.
- the speech recognition technology may refer to a process of converting vocabulary content in human speech into computer readable input; and the inner code refers to a binary character encoding used in a computer Chinese character system, which is a communication input, output, and system platform.
- the handwriting key data corresponding to the voice data can be obtained, thereby correlating the voice data with the user's handwriting handwriting by the inner code, and the vocabulary content in the voice can be converted into the user handwriting handwriting, wherein the voice
- the inner code is essentially the same as the inner code of the handwriting in the handwriting database.
- the handwriting restoration processing can be performed on the key data to obtain the handwritten restoration handwriting; specifically, when the data type of the key data is the trajectory data and/or the vector graphic data, the coordinates of the trajectory coordinate points can be obtained according to the trajectory data.
- Write order connect the coordinate points through the line segment, you can restore the user's handwriting; when the data type of the key data is the screenshot data, you can directly display the screenshot data, or you can analyze the screenshot data. After the corresponding track point coordinates, the line points are connected to the track points, thereby restoring the user's handwriting.
- step S110 may further include the following steps:
- the key data includes any one or any combination of the following data of the handwritten handwriting: track data, screenshot data, and vector graphic data;
- the handwriting recognition process is performed on the handwriting, and the step of obtaining the handwriting code corresponding to the handwritten handwriting may include:
- the step of collecting key data corresponding to the handwriting can include:
- the feature data includes coordinate data of the track point and writing sequence data
- the line points are connected to the track points to obtain a handwritten track
- Corresponding data processing or conversion is performed on the handwritten trajectory to obtain the key data.
- the present invention can track handwritten handwriting input by a user through a handwriting input module (which can include an infrared, a capacitor, and a resistive touch screen device), and obtain a handwritten handwriting; and then perform handwriting recognition processing on the handwriting handwriting by handwriting recognition technology to obtain a handwriting handwriting correspondence.
- the internal code of the handwriting at the same time, the key data corresponding to the handwriting is collected.
- the coordinate position and writing of the "point p(x, y)" when the user's finger or the pen moves can be recorded.
- these "points” can be connected by line segments to form a handwritten trajectory, and the trajectory data, screenshot data or vector graphic data formed by these "dot” data is the key data to be saved by the present invention, and the data processing of the handwritten trajectory is specifically performed. Or data conversion (such as data screenshots, track storage or vector graphics based on handwritten trajectories) to get the corresponding type of key data.
- the handwriting database is established by collecting data of the user's personal handwriting (the handwriting data stored in the database has a personalized feature, so the speed of the internal code retrieval can be improved) Therefore, the handwriting database in the present invention can meet the needs of the user's actual writing habits, and improve the efficiency of processing or converting the user's voice data into personal handwriting.
- the present invention can directly save the "point” data as key data, and simultaneously input the input time stamp of the point (ie, the writing order), and directly connect the "points" by the line segments in the process of handwriting restoration. Thereby the user's handwriting is restored.
- Step S130 may further include the following steps:
- the handwriting restore handwriting is displayed on the display device.
- the user when the user writes on a writable device and encounters a word that does not understand writing, it can be converted into a written handwriting for supplementation (displayed on the display device) by voice input by the method of the present invention, without using other words.
- the way to know how to write and then manually write it up, the process is simple and easy to use.
- the display of the handwriting can also be realized on the writing device by the present invention, thereby improving work efficiency.
- FIG. 2 is a schematic flowchart of Embodiment 2 of a method for converting a handwriting handwriting according to the present invention; as shown in FIG. 2:
- the handwritten handwriting input by the user in the handwriting input device and when the internal code corresponding to the handwritten handwriting can be obtained by handwriting recognition technology (try to shield the invalid or useless handwritten handwriting, save equipment cost, improve recognition efficiency, and ensure conversion by voice)
- Obtaining the accuracy of the handwritten handwriting obtaining the key data of the handwritten handwriting, and storing the inner code and the key data, that is, the handwriting handwriting handwriting data, in the corresponding handwriting database; meanwhile, the user can be repeatedly repeated to obtain the user More handwriting data to ensure the comprehensiveness and accuracy of the data in the handwriting database.
- the voice inner code is retrieved in the handwriting database, and when the handwriting inner code having the same content is retrieved, the handwriting is obtained.
- the key data of the handwritten handwriting corresponding to the code, and according to the key data, the user's handwriting is restored and displayed on the display device.
- the voice inner code obtained by the voice recognition is retrieved in the handwriting database, and the key data corresponding to the voice data is obtained, and then according to The key data restores the user's handwriting, so that the voice data is associated with the user's handwriting by the inner code, and the vocabulary content in the voice can be converted into the user's handwriting; when the user writes on the writable device, the user can pass Voice input, the voice data is converted into writing handwriting for supplement, the process is simple and convenient to use.
- Apparatus for translating handwritten handwriting of the present invention is as follows:
- FIG. 3 is The schematic diagram of the structure of the embodiment 1 of the device for converting the handwritten handwriting of the present invention; as shown in FIG. 3, the method may include:
- the voice recognition unit 310 is configured to perform voice recognition processing on the voice data when the voice data input by the user is received, to obtain a voice inner code corresponding to the voice data;
- the searching unit 320 is configured to retrieve the inner code of the voice in the handwriting database, and obtain the key data of the handwriting hand corresponding to the inner code of the handwriting from the handwriting database when the handwritten inner code corresponding to the inner code of the voice is retrieved; the handwriting restoration unit 330, used for handwriting restoration processing on key data, and obtaining handwritten restoration handwriting.
- the speech internal code obtained by the speech recognition is retrieved in the handwriting database by the retrieval unit 320, and when the corresponding handwriting internal code is retrieved by the retrieval unit 320, The handwriting key data corresponding to the voice data is obtained, thereby associating the voice data with the user's handwriting stroke by the inner code, and the handwriting restoration unit 330 can convert the vocabulary content in the voice into the user handwriting.
- the handwriting handwriting input by the user in the handwriting input device can be obtained by the device for converting the handwriting handwriting according to the present invention, and when the internal code corresponding to the handwritten handwriting can be obtained by the handwriting recognition technology (try to shield the invalid or useless handwritten handwriting, Save equipment costs and improve recognition efficiency, At the same time, the accuracy of the handwritten handwriting obtained by the voice conversion is ensured, and the key data of the handwritten handwriting is obtained, and the inner code and the key data, that is, the handwriting data of the handwritten handwriting are stored in the corresponding handwriting database; Repeat the above steps to get more handwriting data from the user to ensure the comprehensiveness and accuracy of the data in the handwriting database.
- the voice inner code is retrieved in the handwriting database, and when the handwriting inner code having the same content is retrieved, The key data of the handwriting hand corresponding to the inner code of the handwriting, and according to the key data, the handwriting of the user is restored, and the display is performed on the display device.
- the device embodiment 1 for voice conversion handwriting can further include:
- the handwriting recognition unit 340 is configured to track handwriting handwriting input by the user, perform handwriting recognition processing on the handwriting handwriting, and obtain a handwriting internal code corresponding to the handwritten handwriting;
- the key data obtaining unit 350 is configured to collect key data corresponding to the handwriting handwriting;
- the key data includes any one or any combination of the following data of the handwriting handwriting: the track data, the screenshot data, and the vector graphic data;
- the storage unit 360 is configured to store the note internal code and the key data in the handwriting database.
- the key data obtaining unit 350 may include:
- the collecting module 352 is configured to collect feature data of a track point that the user passes when performing handwriting; the feature data includes coordinate data of the track point and writing sequence data;
- the trajectory obtaining module 354 is configured to perform line segment connection on the trajectory points according to the feature data to obtain a handwritten trajectory
- the data processing module 356 is configured to perform corresponding data processing or conversion on the handwritten trajectory to obtain key data.
- the handwriting restoration unit 330 can include:
- the collecting module 332 is configured to acquire feature data of a trajectory that the user passes when performing handwriting according to the key data; the feature data includes coordinate data of the track point and writing sequence data;
- the data processing module 334 is configured to perform line segment connection on the track points according to the feature data, to obtain a handwritten restoration handwriting.
- the present invention can track the handwriting handwriting input by the user through the handwriting recognition unit 340, acquire the handwriting handwriting, and perform handwriting recognition processing on the handwriting handwriting to obtain the handwriting inner code corresponding to the handwritten handwriting.
- the key data acquiring unit 360 collects key data corresponding to the handwriting handwriting. Specifically, when the user writes the input handwriting, the position of the “point p(x, y)” that the user's finger or the pen moves when recording may be recorded, and then These "dot" are connected by line segments to form handwriting, and these "dot” data are the key data to be saved by the present invention (which can also be track data, screenshot data or vector graphic data).
- the apparatus for converting a handwritten handwriting may further include: a display unit 380, configured to display the handwritten restored handwriting on the display device.
- the voice data can be converted into a writing handwriting for supplementation by voice input by the device of the present invention (display)
- the display device On the display device, there is no need to know how to write and then manually write it by other means, the process is simple and convenient to use.
- the device embodiment 1 of the speech conversion handwriting handwriting of the present invention retrieves the inner code obtained by the voice recognition in the handwriting database, and obtains the handwriting key data corresponding to the voice data, thereby associating the voice data with the user's handwriting handwriting by the inner code.
- the vocabulary content in the voice can be converted into the user's handwriting handwriting; when the user writes on the writable device and encounters a word that does not understand writing, it can be converted into a writing handwriting for supplementation by voice input by the method of the present invention, You need to know how to write and then write it manually by other means.
- the process is simple and easy to use.
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
一种语音转化手写笔迹的方法和装置，语音转化手写笔迹的方法，包括以下步骤：在接收到用户输入的语音数据时，对语音数据进行语音识别处理，得到语音数据对应的语音内码（S110）；在笔迹数据库中对语音内码进行检索，在检索到与语音内码对应的笔迹内码时，从笔迹数据库中获取笔迹内码对应的手写笔迹的关键数据（S120）；关键数据包括手写笔迹的以下数据中的任意一种或任意组合：轨迹数据、截图数据和矢量图形数据；对关键数据进行笔迹还原处理，得到手写还原笔迹（S130）。使得用户在可书写设备上进行书写时，可以通过语音输入，将语音数据转化成书写笔迹进行补充，过程简单，使用方便。
Description
本发明涉及信息技术领域，特别是涉及一种语音转化手写笔迹的方法和装置。
当前随着智能手机、掌上电脑、交互式电子白板等信息工具的普及，以及智能语音合成、语音识别技术的发展，传统的语音技术已经无法满足用户的使用要求。
在实现过程中，发明人发现传统技术中至少存在如下问题：传统的语音技术无法将语音转化成用户手写笔迹，影响输入效率：例如，当用户在可书写设备上进行书写，遇到不懂书写的字词(即忘记或不知字词的书写笔画和拼写方式)时，需要通过其他查找途径得到具体的书写笔画或拼写方式，再手动写到书写设备上去，整个过程操作繁琐，不够便利；或当用户双手被占用，无法在书写设备上及时完成手写操作时，会直接影响工作效率。
发明内容
基于此，有必要针对传统的语音技术无法将语音转化成用户手写笔迹的问题，提供一种语音转化手写笔迹的方法和装置。
为了实现上述目的，本发明技术方案的实施例为：
一方面，提供了一种语音转化手写笔迹的方法，包括以下步骤：
在接收到用户输入的语音数据时，对语音数据进行语音识别处理，得到
语音数据对应的语音内码；
在笔迹数据库中对语音内码进行检索，在检索到与语音内码对应的笔迹内码时，从笔迹数据库中获取笔迹内码对应的手写笔迹的关键数据；
对关键数据进行笔迹还原处理，得到手写还原笔迹。
另一方面，提供了一种语音转化手写笔迹的装置，包括：
语音识别单元，用于在接收到用户输入的语音数据时，对语音数据进行语音识别处理，得到语音数据对应的语音内码；
检索单元，用于在笔迹数据库中对语音内码进行检索，在检索到与语音内码对应的笔迹内码时，从笔迹数据库中获取笔迹内码对应的手写笔迹的关键数据；
笔迹还原单元，用于对关键数据进行笔迹还原处理，得到手写还原笔迹。
上述技术方案具有如下有益效果：
本发明语音转化手写笔迹的方法和装置，在笔迹数据库中对语音识别得到的内码进行检索，得到语音数据所对应的笔迹关键数据，从而通过内码将语音数据和用户的手写笔迹关联起来，可以将语音中的词汇内容转化成用户手写笔迹；当用户在可书写设备上进行书写时，可以通过语音输入，将语音数据转化成书写笔迹进行补充，过程简单，使用方便。
图1为本发明语音转化手写笔迹的方法实施例1的流程示意图；
图2为本发明语音转化手写笔迹的方法实施例2的流程示意图；
图3为本发明语音转化手写笔迹的装置实施例1的结构示意图。
为了便于理解本发明，下面将参照相关附图对本发明进行更全面的描述。附图中给出了本发明的首选实施例。但是，本发明可以以许多不同的形式来实现，并不限于本文所描述的实施例。相反地，提供这些实施例的目的是使对本发明的公开内容更加透彻全面。
除非另有定义，本文所使用的所有的技术和科学术语与属于本发明的技术领域的技术人员通常理解的含义相同。本文中在本发明的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本发明。本文所使用的术语“及/或”包括一个或多个相关的所列项目的任意的和所有的组合。
为了解决传统的语音技术无法将语音转化成用户手写笔迹的问题，本发明提供了一种语音转化手写笔迹的方法实施例1，图1为本发明语音转化手写笔迹的方法实施例1的流程示意图；如图1所示，语音转化手写笔迹的方法实施例1可以包括以下步骤：
步骤S110：在接收到用户输入的语音数据时，对语音数据进行语音识别处理，得到语音数据对应的语音内码；
步骤S120：在笔迹数据库中对语音内码进行检索，在检索到与语音内码对应的笔迹内码时，从笔迹数据库中获取笔迹内码对应的手写笔迹的关键数据；
步骤S130：对关键数据进行笔迹还原处理，得到手写还原笔迹。
具体而言，语音识别技术可以指将人类的语音中的词汇内容转换为计算机可读的输入的过程；而内码是指计算机汉字系统中使用的二进制字符编码，是沟通输入、输出与系统平台之间的交换码；当接收用户手写输入模块在预设时间段内未接收到相应的手写输入数据时，设备可确定用户不懂如何书写
或双手被占用，可以弹出语音输入框选项或直接激活语音输入模块，从而可以利用本发明语音转化手写笔迹的方法实施例1，在笔迹数据库中对语音识别得到的语音内码进行检索，在检索到相应的笔迹内码时，即可得到语音数据所对应的笔迹关键数据，从而通过内码将语音数据和用户的手写笔迹关联起来，可以将语音中的词汇内容转化成用户手写笔迹，其中语音内码与笔迹数据库中的笔迹内码的实质是相同的。
进一步的，可以对关键数据进行笔迹还原处理，从而得到手写还原笔迹；具体而言，当关键数据的数据类型为轨迹数据和/或矢量图形数据时，可以根据轨迹数据获取轨迹坐标点的坐标和写入顺序，通过线段将坐标点连接起来，即可还原出用户手写笔迹；当关键数据的数据类型为截图数据时，既可以直接对截图数据进行显示，也可以对截图数据进行解析，在获取到相应的轨迹点坐标后，对轨迹点进行线段连接，从而还原出用户的手写笔迹。
在一个具体的实施例中，步骤S110之前还可以包括步骤：
跟踪用户输入的手写笔迹，对手写笔迹进行手写识别处理，得到手写笔迹对应的笔迹内码；
采集手写笔迹对应的关键数据；关键数据包括手写笔迹的以下数据中的任意一种或任意组合：轨迹数据、截图数据和矢量图形数据；
将笔记内码和关键数据存储在笔迹数据库中。
在一个具体的实施例中对手写笔迹进行手写识别处理，得到手写笔迹对应的笔迹内码的步骤可以包括：
获取手写笔迹的有序轨迹信息；
对有序轨迹信息进行内码转换，得到笔迹内码。
在一个具体的实施例中，采集手写笔迹对应的关键数据的步骤可以包括：
采集用户在进行手写时经过的轨迹点的特征数据；特征数据包括轨迹点的坐标数据和写入顺序数据；
根据特征数据，对轨迹点进行线段连接，得到手写轨迹；
对手写轨迹进行相应的数据处理或转换，得到所述关键数据。
具体而言，本发明可以通过手写输入模块(可以包括红外、电容、电阻触摸屏设备)跟踪用户输入的手写笔迹，获取手写笔迹；然后通过手写识别技术对手写笔迹进行手写识别处理，得到手写笔迹对应的笔迹内码；同时采集手写笔迹对应的关键数据，具体而言，在用户书写输入笔迹时，可以记录用户手指或者笔移动时经过的“点p(x，y)”的坐标位置和写入顺序，然后可以用线段将这些“点”连接起来形成手写轨迹，而这些“点”数据形成的轨迹数据、截图数据或矢量图形数据就是本发明要保存的关键数据，具体对手写轨迹进行数据处理或数据转换(例如数据截图、轨迹存储或根据手写轨迹进行矢量图形的绘制)，从而得到相应类型的关键数据。
采用这种关键数据的采集方式，在用户在使用书写设备进行手写时，可以完全还原出用户的手写笔迹，保证笔迹还原的准确性。此外，在本发明的各实施例中，是通过对用户个人的手写笔迹进行数据采集，建立起的笔迹数据库(数据库中保存的笔迹数据具有个性化的特点，因此可以提高内码检索的速度)，使得本发明中的笔迹数据库能够贴合用户的实际书写习惯的需要，提高将用户的语音数据处理或转换成个人的手写笔迹的效率。
进一步的，本发明也可直接将“点”数据作为关键数据进行保存，同时记录点的输入时间戳(即写入顺序)，在笔迹还原的过程中直接用线段对“点”进行依次连接，从而还原出用户的手写笔迹。
步骤S130之后还可以包括步骤：
将手写还原笔迹在显示设备上进行显示。
具体而言，当用户在可书写设备上书写，遇到不懂书写的字词时，可以通过语音输入，由本发明的方法转化成书写笔迹进行补充(显示在显示设备上)，不需要通过其他途径知道怎么书写后再手动写上去，过程简单，使用方便。当用户双手被占用时，也可以通过本发明在书写设备上实现手写笔迹的显示，从而提高工作效率。
本发明语音转化手写笔迹的方法实施例2：
为了解决传统的语音技术无法将语音转化成用户手写笔迹的问题，同时详细阐述本发明的技术方案，特以具体应用本发明语音转化手写笔迹的方法进行语音转化的实例为例，说明本发明的实现过程，图2为本发明语音转化手写笔迹的方法实施例2的流程示意图；如图2所示：
获取用户在手写输入设备中输入的手写笔迹，在可以通过手写识别技术获取手写笔迹对应的内码时(尽量屏蔽掉无效或无用的手写笔迹，节约设备成本，提高识别效率，同时保证由语音转化得到的手写笔迹的准确性)，获取该手写笔迹的关键数据，并将内码和关键数据即该手写笔迹的笔迹数据存储在相应的笔迹数据库中；同时，可以通过不断重复上述步骤，得到用户更多的笔迹数据，以保证笔迹数据库中数据的全面性与准确性。
获取用户输入的语音数据，在可以通过语音识别技术获取语音数据对应的内码时，在上述笔迹数据库中对语音内码进行检索，到检索到具有相同内容的笔迹内码时，获取该笔迹内码对应的手写笔迹的关键数据，并根据关键数据，还原出用户笔迹，同时在显示设备上进行显示。
本发明语音转化手写笔迹的方法的各实施例，在笔迹数据库中对语音识别得到的语音内码进行检索，得到语音数据所对应的笔迹关键数据，再根据
关键数据还原出用户的手写笔迹，从而通过内码将语音数据和用户的手写笔迹关联起来，可以将语音中的词汇内容转化成用户手写笔迹；当用户在可书写设备上进行书写时，可以通过语音输入，将语音数据转化成书写笔迹进行补充，过程简单，使用方便。
本发明语音转化手写笔迹的装置实施例1：
基于以上语音转化手写笔迹的方法的技术思想，同时为了解决传统的语音技术无法将语音转化成用户手写笔迹的问题，本发明还提供了一种语音转化手写笔迹的装置实施例1，图3为本发明语音转化手写笔迹的装置实施例1的结构示意图；如图3所示，可以包括：
语音识别单元310，用于在接收到用户输入的语音数据时，对语音数据进行语音识别处理，得到语音数据对应的语音内码；
检索单元320，用于在笔迹数据库中对语音内码进行检索，在检索到与语音内码对应的笔迹内码时，从笔迹数据库中获取笔迹内码对应的手写笔迹的关键数据；笔迹还原单元330，用于对关键数据进行笔迹还原处理，得到手写还原笔迹。
具体而言，本发明语音转化手写笔迹的装置实施例1，通过检索单元320在笔迹数据库中对语音识别得到的语音内码进行检索，在检索单元320检索到相应的笔迹内码时，即可得到语音数据所对应的笔迹关键数据，从而通过内码将语音数据和用户的手写笔迹关联起来，笔迹还原单元330可以将语音中的词汇内容转化成用户手写笔迹。
进一步的，可以通过本发明语音转化手写笔迹的装置获取用户在手写输入设备中输入的手写笔迹，在可以通过手写识别技术获取手写笔迹对应的内码时(尽量屏蔽掉无效或无用的手写笔迹，节约设备成本，提高识别效率，
同时保证由语音转化得到的手写笔迹的准确性)，通过获取该手写笔迹的关键数据，并将内码和关键数据即该手写笔迹的笔迹数据存储在相应的笔迹数据库中；同时，可以通过不断重复上述步骤，得到用户更多的笔迹数据，以保证笔迹数据库中数据的全面性与准确性。
此外接收到用户输入的语音数据后，在可以通过语音识别技术获取语音数据对应的内码时，在上述笔迹数据库中对语音内码进行检索，到检索到具有相同内容的笔迹内码时，获取该笔迹内码对应的手写笔迹的关键数据，并根据该关键数据，还原出用户笔迹，同时在显示设备上进行显示。
在一个具体的实施例中，语音转化手写笔迹的装置实施例1还可以包括：
手写识别单元340，用于跟踪用户输入的手写笔迹，对手写笔迹进行手写识别处理，得到手写笔迹对应的笔迹内码；
关键数据获取单元350，用于采集手写笔迹对应的关键数据；关键数据包括手写笔迹的以下数据中的任意一种或任意组合：轨迹数据、截图数据和矢量图形数据；
存储单元360，用于将笔记内码和关键数据存储在笔迹数据库中。
在一个具体的实施例中，关键数据获取单元350可以包括：
采集模块352，用于采集用户在进行手写时经过的轨迹点的特征数据；特征数据包括轨迹点的坐标数据和写入顺序数据；
轨迹获取模块354，用于根据特征数据，对轨迹点进行线段连接，得到手写轨迹；
数据处理模块356，用于对手写轨迹进行相应的数据处理或转换，得到关键数据。
在一个具体的实施例中，笔迹还原单元330可以包括：
采集模块332，用于根据关键数据，获取用户在进行手写时经过的轨迹的特征数据；特征数据包括轨迹点的坐标数据和写入顺序数据；
数据处理模块334，用于根据特征数据，对轨迹点进行线段连接，得到手写还原笔迹。
具体而言，本发明可以通过手写识别单元340跟踪用户输入的手写笔迹，获取手写笔迹，并对手写笔迹进行手写识别处理，得到手写笔迹对应的笔迹内码。并通过关键数据获取单元360采集手写笔迹对应的关键数据，具体而言，在用户书写输入笔迹时，可以记录用户手指或者笔移动时经过的“点p(x，y)”的位置，然后可以用线段将这些“点”连接起来形成笔迹，那么这些“点”数据就是本发明要保存的关键数据(也可以为轨迹数据、截图数据或矢量图形数据)。
在一个具体的实施例中，语音转化手写笔迹的装置还可以包括：显示单元380，用于将手写还原笔迹在显示设备上进行显示。
具体而言，当用户在可书写设备上书写，遇到不懂书写的字词时或者双手被占用的情况时，可以通过语音输入，由本发明的装置将语音数据转化成书写笔迹进行补充(显示在显示设备上)，不需要通过其他途径知道怎么书写后再手动写上去，过程简单，使用方便。
本发明语音转化手写笔迹的装置实施例1，在笔迹数据库中对语音识别得到的内码进行检索，得到语音数据所对应的笔迹关键数据，从而通过内码将语音数据和用户的手写笔迹关联起来，可以将语音中的词汇内容转化成用户手写笔迹；当用户在可书写设备上书写，遇到不懂书写的字词时，可以通过语音输入，由本发明的方法转化成书写笔迹进行补充，不需要通过其他途径知道怎么书写后再手动写上去，过程简单，使用方便。
以上所述实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本发明的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干变形和改进，这些都属于本发明的保护范围。因此，本发明专利的保护范围应以所附权利要求为准。
Claims (10)
- 一种语音转化手写笔迹的方法，其特征在于，包括以下步骤：在接收到用户输入的语音数据时，对所述语音数据进行语音识别处理，得到所述语音数据对应的语音内码；在笔迹数据库中对所述语音内码进行检索，在检索到与所述语音内码对应的笔迹内码时，从所述笔迹数据库中获取所述笔迹内码对应的手写笔迹的关键数据；对所述关键数据进行笔迹还原处理，得到手写还原笔迹。
- 根据权利要求1所述的语音转化手写笔迹的方法，其特征在于，在接收到用户输入的语音数据时，对所述语音数据进行语音识别处理，得到所述语音数据对应的语音内码的步骤之前还包括步骤：跟踪用户输入的手写笔迹，对所述手写笔迹进行手写识别处理，得到所述手写笔迹对应的所述笔迹内码；采集所述手写笔迹对应的所述关键数据；所述关键数据包括所述手写笔迹的以下数据中的任意一种或任意组合：轨迹数据、截图数据和矢量图形数据；将所述笔记内码和所述关键数据存储在所述笔迹数据库中。
- 根据权利要求2所述的语音转化手写笔迹的方法，其特征在于，对所述手写笔迹进行手写识别处理，得到所述手写笔迹对应的所述笔迹内码的步骤包括：获取所述手写笔迹的有序轨迹信息；对所述有序轨迹信息进行内码转换，得到所述笔迹内码。
- 根据权利要求2所述的语音转化手写笔迹的方法，其特征在于，采集 所述手写笔迹对应的所述关键数据的步骤包括：采集用户在进行手写时经过的轨迹点的特征数据；所述特征数据包括所述轨迹点的坐标数据和写入顺序数据；根据所述特征数据，对所述轨迹点进行线段连接，得到手写轨迹；对所述手写轨迹进行相应的数据处理或转换，得到所述关键数据。
- 根据权利要求1至4任意一项所述的语音转化手写笔迹的方法，其特征在于，对所述关键数据进行笔迹还原处理，得到手写还原笔迹的步骤之后还包括步骤：将所述手写还原笔迹在显示设备上进行显示。
- 一种语音转化手写笔迹的装置，其特征在于，包括：语音识别单元，用于在接收到用户输入的语音数据时，对所述语音数据进行语音识别处理，得到所述语音数据对应的语音内码；检索单元，用于在笔迹数据库中对所述语音内码进行检索，在检索到与所述语音内码对应的笔迹内码时，从所述笔迹数据库中获取所述笔迹内码对应的手写笔迹的关键数据；笔迹还原单元，用于对所述关键数据进行笔迹还原处理，得到手写还原笔迹。
- 根据权利要求6所述的语音转化手写笔迹的装置，其特征在于，还包括：手写识别单元，用于跟踪用户输入的手写笔迹，对所述手写笔迹进行手写识别处理，得到所述手写笔迹对应的所述笔迹内码；关键数据获取单元，用于采集所述手写笔迹对应的所述关键数据；所述关键数据包括所述手写笔迹的以下数据中的任意一种或任意组合：轨迹数据、 截图数据和矢量图形数据；存储单元，用于将所述笔记内码和所述关键数据存储在所述笔迹数据库中。
- 根据权利要求7所述的语音转化手写笔迹的装置，其特征在于，所述手写识别单元包括：轨迹信息获取模块，用于获取所述手写笔迹的有序轨迹信息；内码转换模块，用于对所述有序轨迹信息进行内码转换，得到所述笔迹内码。
- 根据权利要求7所述的语音转化手写笔迹的装置，其特征在于，所述关键数据获取单元包括：采集模块，用于采集用户在进行手写时经过的轨迹点的特征数据；所述特征数据包括所述轨迹点的坐标数据和写入顺序数据；轨迹获取模块，用于根据所述特征数据，对所述轨迹点进行线段连接，得到手写轨迹；数据处理模块，用于对所述手写轨迹进行相应的数据处理或转换，得到所述关键数据。
- 根据权利要求6至9任意一项所述的语音转化手写笔迹的装置，其特征在于，所述语音转化手写笔迹的装置还包括：显示单元，用于将所述手写还原笔迹在显示设备上进行显示。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610571390.7 | 2016-07-18 | ||
| CN201610571390.7A CN106205632B (zh) | 2016-07-18 | 2016-07-18 | 语音转化手写笔迹的方法和装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2018014570A1 true WO2018014570A1 (zh) | 2018-01-25 |
Family
ID=57494562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2017/077049 WO2018014570A1 (zh) | 2016-07-18 | 2017-03-17 | 语音转化手写笔迹的方法和装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106205632B (zh) |
| WO (1) | WO2018014570A1 (zh) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106205632B (zh) * | 2016-07-18 | 2019-07-09 | 广州视睿电子科技有限公司 | 语音转化手写笔迹的方法和装置 |
| CN108363522A (zh) * | 2018-04-24 | 2018-08-03 | 石家庄科达文教用品有限公司 | 同步书写系统及其方法 |
| CN110970033A (zh) * | 2019-12-02 | 2020-04-07 | 浙江理工大学 | 一种声纹识别自动签字装置及其识别方法 |
| CN110868427A (zh) * | 2019-12-02 | 2020-03-06 | 浙江理工大学 | 一种远程语音识别自动签字装置及其识别方法 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1549244A (zh) * | 2003-05-22 | 2004-11-24 | 宏�股份有限公司 | 整合语音与手写输入的识别方法及其系统 |
| CN1737902A (zh) * | 2005-09-12 | 2006-02-22 | 周运南 | 文字语音互转装置 |
| CN101315666A (zh) * | 2008-07-11 | 2008-12-03 | 中国科学院软件研究所 | 一种基于语音的多通道手写中文纠错方法 |
| JP4407494B2 (ja) * | 2004-11-26 | 2010-02-03 | 株式会社日立製作所 | デジタルペンを用いた速記文字反訳作業支援システム |
| CN105046730A (zh) * | 2015-07-09 | 2015-11-11 | 北京盛世宣合信息科技有限公司 | 应用于毛笔的书写笔迹呈现方法和装置 |
| CN106205632A (zh) * | 2016-07-18 | 2016-12-07 | 广州视睿电子科技有限公司 | 语音转化手写笔迹的方法和装置 |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101146292A (zh) * | 2007-10-26 | 2008-03-19 | 中兴通讯股份有限公司 | 一种语音编辑短信息的方法及手机装置 |
| CN101763360A (zh) * | 2008-12-08 | 2010-06-30 | 新奥特硅谷视频技术有限责任公司 | 一种庭审过程中语音信息转换成文字信息的方法和装置 |
| CN101877839A (zh) * | 2010-05-25 | 2010-11-03 | 南京邮电大学 | 一种个性化短消息传输方法 |
| KR20120010819A (ko) * | 2010-07-27 | 2012-02-06 | 주식회사 파인디지털 | 음성 인식 기능을 갖는 정보 단말기 및 이를 위한 음성 명령어 구조 |
| JP5660441B2 (ja) * | 2010-09-22 | 2015-01-28 | 独立行政法人情報通信研究機構 | 音声認識装置、音声認識方法、及びプログラム |
| CN102135821B (zh) * | 2011-03-08 | 2013-03-06 | 中国科学技术大学 | 一种手写笔及一种图形还原系统 |
| CN202276392U (zh) * | 2011-10-21 | 2012-06-13 | 宁波波导软件有限公司 | 一种能够保持手写真迹的手机 |
| CN105159870B (zh) * | 2015-06-26 | 2018-06-29 | 徐信 | 一种精准完成连续自然语音文本化的处理系统及方法 |
-
2016
- 2016-07-18 CN CN201610571390.7A patent/CN106205632B/zh active Active
-
2017
- 2017-03-17 WO PCT/CN2017/077049 patent/WO2018014570A1/zh active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1549244A (zh) * | 2003-05-22 | 2004-11-24 | 宏�股份有限公司 | 整合语音与手写输入的识别方法及其系统 |
| JP4407494B2 (ja) * | 2004-11-26 | 2010-02-03 | 株式会社日立製作所 | デジタルペンを用いた速記文字反訳作業支援システム |
| CN1737902A (zh) * | 2005-09-12 | 2006-02-22 | 周运南 | 文字语音互转装置 |
| CN101315666A (zh) * | 2008-07-11 | 2008-12-03 | 中国科学院软件研究所 | 一种基于语音的多通道手写中文纠错方法 |
| CN105046730A (zh) * | 2015-07-09 | 2015-11-11 | 北京盛世宣合信息科技有限公司 | 应用于毛笔的书写笔迹呈现方法和装置 |
| CN106205632A (zh) * | 2016-07-18 | 2016-12-07 | 广州视睿电子科技有限公司 | 语音转化手写笔迹的方法和装置 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106205632B (zh) | 2019-07-09 |
| CN106205632A (zh) | 2016-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2018014570A1 (zh) | 语音转化手写笔迹的方法和装置 | |
| KR101825154B1 (ko) | 중첩 쓰기 핸드라이팅 입력방법 | |
| CN100587660C (zh) | 一种手写字符预测识别的方法和装置 | |
| US8364680B2 (en) | Computer systems and methods for collecting, associating, and/or retrieving data | |
| CN103502915A (zh) | 电子设备上基于屏幕软件健盘实现滑动输入文本的系统及方法 | |
| CN101465074A (zh) | 一种手写汉字布局评价方法 | |
| JP6464504B2 (ja) | 電子機器、処理方法およびプログラム | |
| CN102930270A (zh) | 基于肤色检测与背景消除的手部识别方法及系统 | |
| CN109002803A (zh) | 一种基于智能手表的握笔姿势检测和汉字笔顺识别方法 | |
| CN110930129B (zh) | 结合周工作计划的点阵纸笔技术应用方法 | |
| CN115497102A (zh) | 一种甲骨文识别方法及系统 | |
| CN106354260B (zh) | 一种搜题的方法及装置 | |
| CN110532415A (zh) | 图像搜索处理方法、装置、设备及存储介质 | |
| CN103412677A (zh) | 一种手绘内容识别的方法及装置 | |
| CN102929394A (zh) | 一种基于手势识别的盲文输入法 | |
| CN103123750A (zh) | 结合语言学习功能的数字相框及其显示方法 | |
| CN110910291A (zh) | 结合康奈尔或东大笔记法的点阵纸笔技术应用方法 | |
| CN110910290A (zh) | 基于点阵笔技术进行错题管理的方法 | |
| CN1996217A (zh) | 基于手写输入方式的转换输入设备和方法 | |
| CN110990535B (zh) | 基于点阵纸笔技术进行会议记录的方法 | |
| CN102521577A (zh) | 一种交互式多媒体设备的笔迹识别、合成和跟踪方法 | |
| CN117648298B (zh) | 一种平板电脑智能管理系统、方法及装置 | |
| CN102566755A (zh) | 一种繁简对照学习的输入装置和方法 | |
| KR20120127000A (ko) | 문자인식을 이용한 이미지 관리 장치 및 그 방법 | |
| US10248214B2 (en) | Wearable device for writing and drawing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17830213 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 17830213 Country of ref document: EP Kind code of ref document: A1 |