SinoVoice has acquired many core technologies in the areas of TTS (Text-to-Speech), ASR (Automatic Speech Recognition) and HWR (Handwriting Recognition) and has become one of the few companies possessing multiple HCI (Human Computer Interaction) technologies in China, due to its years of dedication to the research of HCI. In August 2006, SinoVoice established the Human Computer Interface center (HCI center), which gives the company a platform to integrate, combine and productize its core technologies.
1. Scopes of application
It can be applied on GPS/GIS terminals, including:Personal terminal, vehicle terminals
2. Application software products
1) News broadcast
Functions:Broadcast the driving route of destinations, provide real-time traffic information, weather forecast and speed reminder in voice.
Traffic information broadcast:The system will send the traffic information to the vehicle GPS screen in the form of text and read it out at the same time. The driver only needs to concentrate on the steering wheel and change the driving route according to the information.
Weather forecast: the weather has been more and more important for the drivers. The weather is most changeable in different time and different places. We may tell our family that it is raining in our place, while it may be sunny at home. Prepare for bad weather in advance can not only protect your precious car, but give you better secure assurance which is more important.
Reminding function:This function includes speed reminding, forward traffic reminding, route plan reminding, signal status reminding, accident broadcast, etc. In the process of driving, if you forget change driving line or miss an exit, which is normal for new drivers, the GPS devices will remind you, ‘ the exit is ahead. Please change your driving line in time.’ When young people drive on the road, they may exceed speed limits unconsciously. The GPS devices will remind you, ‘The car has exceeded the speed limit. Please drive safe.’ The GPS devices can help you avoid driving mistakes.
2) Handwriting input
Function:Input gas station, destination and other search information by handwriting.
SinoVoice’s HWR technology can make us write the destination in the shortest time, avoiding the complicated key tapping operation. For example, if you want to know how to get to the nearest gas station, you can write ‘gas station’ in the text input block. The system will point out the location of the nearby gas stations in the electronic map.
3) Voice command navigation
Function:Enter any menu of the GPS device or execute related command by voice, or input destination, nearby hotels and other search information by voice.
For example:
“Please tell me your location”
“5th street of Shang Di”
“Where are you heading to? ”
“Wang Fu jing”
The above is the conversation between the driver and the GPS system. When the recognition system knows where you want to go, it will list the best route for you on the screen. It will make your life much easier.
3. Core technology
1) Handwriting recognition technology
Handwriting recognition is the process of transferring orderly trail information, which is formed while the user writs on a writing device, into Chinese in-code. It actually is a mapping process from coordinate sequence of writing trails to Chinese in-code.
SinoVoice’s HWR technology which deploys the grammar self-learning algorithm and integrated technology based on characteristics statistics algorithm can recognize Chinese (Simplified Chinese 6763+ Traditional Chinese 5401+ GBK21003), numbers, English and other special symbols. It supports the recognition of casually written characters and characters that are written in wrong orders. The recognition accuracy amounts to over 99%. The recognition is 0.4 second/character~0.6 second/character under MIPS 33M circumstances. The size of the data dictionary can be extended or compressed. It support the transplant on all kinds of embedded platforms.
2) Text-to-Speech technology
TTS is also called Text-to-Speech,which is a technology to transfer any combination of text files into audio files by computer and plays the voice by sound card, telephone card and other multimedia apparatus. Simply put it, TTS can make the machine read out the text files.
SinoVoice’s TTS is a core technology formed by ‘adding voice library compression’ algorithm and ‘voice tone changing’ algorithm. The accuracy of the system amounts to 96.59%, the clarity of syllables is over 94%, the 97.83% of the sentences are understandable,and it sounds as natural as human voice. It supports mix reading of Chinese and English. It provides voice library in different timbres and different languages, such as female, male, female Taiwanese, female Cantonese, etc. the size of the voice library can be extended or compressed, the little voice library is just 400K--4MB large.
3) Automatic Speech Recognition
Automatic Speech Recognition is the technology that can makes machine recognizing and understanding audio information and convert it into correspondent files or command.
The accuracy of SinoVoice’s ASR technology can reach 96% within 200 words and can support recognition for 1000 words. It supports Windows, Linux and other operation systems.
4. Corporation modes
SinoVoice values and respects its clients. It has worked hard to develop cooperative partners to extend business together and get mutual benefit. In the authorization mode of software, there are Quantitative authorization, yearly authorization, Outright Authorization and other cooperative modes. We have different price systems for different cooperative modes, taking full consideration of the technical characteristics of products, the market life cycle and other elements to protect the mutual interests of both parties, ensuring high efficiency and integrity.
Provide SDK for the three core technologies: Embedded Handwriting Recognition – eJHWR、 Embedded Text-to-speech – eJTTS、 Embedded Automatic Speech Recognition – eJASR,and can transplant it onto mainstream platforms such as TI, SpreadTrum, MTK, Philips and other platforms.
We provide Handwriting input, voice command navigation and other GPS application software
We provide personalized GPS application software for customers based on the three core technologies as eJHWR、eJTTS and eJASR, of which SinoVoice has independent property rights.
For example, based on the embedded handwriting recognition technology – eJHWR, we can make customized recognizable word library for the customer, add new words and gesture symbols, and compress the recognition library to meet customer’s demand for the memory and recognition speed, or expand the recognition library to meet customer’s demand for recognition rate.
5. Contact
Beijing InfoQuick SinoVoice Speech Technology Corporation
Address :206 10th building of ZPark, HaiDian District, Beijing, China
Post code:100094
Phone :+86-10-82826886
Fax :+86-10-82825830
Website
捷通华声语音技术有限公司
Email :zhaomiao@sinovoice.com.cn
marketing@sinovoice.com.cn
More Contact,More Possibility!