With the acceleration of urban construction, urban public transportation has also developed rapidly. There are more and more bus routes, which greatly facilitates citizens and tourists. At the same time, bus companies also provide people with a service hotline for telephone inquiry of bus routes. The current service hotline has two modes: manual inquiry and automatic inquiry.
When it comes to manual inquiries, it takes a long time and the results are not good; when it comes to automatic inquiries, people follow the voice prompts and use key operations, and it takes a long time to complete the inquiry. For example: if you want to inquire about the operation status of bus No. 105, you need to press the three numbers "1", "0", and "5" on the phone; if you want to inquire about all bus trips to a certain place, you can only do it manually, because this kind of inquiry service cannot be achieved by pressing keys.
With the significant breakthroughs in computer technology and digital electronic technology, speech recognition technology has also matured, especially in the recognition of large vocabulary and non-specific continuous speech, which has made significant progress and has entered the practical stage [1]. Therefore, we proposed a solution to transform the original service hotline by using a public speech recognition system.
1. Characteristics of bus route intelligent query system
The bus route voice query system includes two parts: key query (only includes bus route query) and voice query (including query by bus route and query by city location). The key query is mainly for old customers, but the main body of the system is voice query, which has the following features:
(1) Interactive question-and-answer query is implemented. Users only need to answer the system prompts to get the information they need.
(2) As for the system itself, it has achieved the popularization of speech recognition. Any user who uses this system for the first time can obtain the required information satisfactorily without undergoing speech training.
(3) The system’s daily call volume can be counted.
2 System Structure and Software Method
The system [1.2] is mainly composed of computer network, artificial seats, voice processing and storage, and system management (database management + voice editing), as shown in Figure 1. This system uses industrial machines with stable performance. The system platform is Windows NT operating system. Its function is to use computers and digital voice cards to process a large number of complex telephone services and provide open bus route intelligent query services. It has the characteristics of fast speed, all-weather, and high efficiency.
2.1 System hardware and software platform configuration and development tool selection (based on 30-line configuration)
The system hardware uses industrial computers to achieve a high performance-price ratio, and the performance of the entire system can be greatly improved through the computer cluster system. Specifically, the CPU is P4 2.4G or above, the memory is 1G or above, and the hard disk is 60G or above.
The system software platform is Windows NT, which has good graphics management functions and provides reliable technical support and development environment.
The voice card uses D321/641 voice card, which is very mature in development under Windows NT and can handle various different needs.
The D321/641[3] voice card has the following features:
(1) It integrates 30/60-channel voice processing and El/2e1 (PCM) digital trunk interface, and supports both SSL and DSL signaling (DSSL).
(2) It is powerful and cost-effective, making it easy for users to build stable and multi-functional application systems.
(3) The full DSP structure has powerful processing capabilities, which can reduce the burden on the host and ensure the efficient and stable operation of high-density, large-capacity application systems. At the same time, DSP technology facilitates software upgrades.
The database system uses SQL Server, which is a relatively mature database product under Windows NT environment. It has all the features of a large database system and is easy to integrate with NT for management.
The development tools include Visual C++, Visual Basic, etc. Visual C++ is powerful and flexible, and can handle various programming needs; Visual Basic is fast in programming and does not require high requirements from programmers, so the two are used together.
2.2 Software Design
The system interface is implemented by Visual Basic 6.0 programming. Its functions are clearly displayed in Chinese characters and icons, which is easy to operate and clear at a glance, making it convenient for on-duty personnel to operate and monitor.
Each functional program is implemented by Visual C++ programming, which is mainly divided into: detecting ringing events and accepting user voice to implement corresponding function calls; determining the effective state of the interface menu after entering the system and detecting and initializing the telephone voice card; logging into the database server.
2.3 Design of database server
The client/server structure [4] is a widely used structure for database-based applications. It takes advantage of the imbalance in the distribution of computer system hardware and software, concentrates data services on the server side, and the recognition client is responsible for processing application logic and interface. This greatly reduces the complexity of program development. At the same time, the recognition client and recognition server communicate and work together in real time, reducing latency and ensuring high recognition efficiency. The client/server structure consists of two independent implementations.
The recognition client and the recognition server. The application running on the former is called the "front end" or "foreground", and the latter is called the "back end". In other words, the server adds a "back-end database engine". When a call is connected, the recognition client takes over the voice input of the phone user, transmits it to the recognition server in real time, and returns the recognition result of the recognition server to the phone application platform layer, and then performs internal calculations to find data that meets the conditions. At this time, the client can also use this idle time to handle other tasks. When the server finishes processing, it will pass the required data to the front-end application.
Since the front-end application of the client/server structure does not directly process the database, ODBC is used as a bridge to help us open the channel between the front-end application and the back-end database management system, and the ODBC API provides standardized control functions to maintain the server's database data.
The speech synthesis process is handled by synthesizing and playing at the same time, avoiding the large delay caused by synthesizing long text. When the system plays the speech in segments, any interruption by the user can terminate the synthesis task, avoiding unnecessary resource consumption.
3 System Implementation
When the system starts, it will first access the DataSource.ini file and read the relevant database information from it. If the data of the database server is changed, only this file needs to be changed. The content of the DataSource.ini file is as follows:
[Source]
namc=flight
[DSN]
name="ODBC;DSN=flight;UID=flight;PWD=flight' Where: name=flight is the data source name, DSN is the data source name, UID is the user ID, and PWD is the user password.
FIG. 2 is a flow chart of the automatic query system implemented by the user through voice.
4 Conclusion
This system is based on the communication system and aims to meet actual needs. It is an easy-to-use and fully functional voice query system designed by combining existing mature hardware products with software development. It integrates voice recognition and speech synthesis technology, changes the old way of using telephone buttons to query, supports voice input, and the humanized voice interface makes query more convenient and more friendly, providing reliable technical support for the hotline service of the public transportation system.
Previous article:Smart Agriculture Wireless Monitoring System Solution
Next article:How to adjust the satellite dish
- Popular Resources
- Popular amplifiers
- High signal-to-noise ratio MEMS microphone drives artificial intelligence interaction
- Advantages of using a differential-to-single-ended RF amplifier in a transmit signal chain design
- ON Semiconductor CEO Appears at Munich Electronica Show and Launches Treo Platform
- ON Semiconductor Launches Industry-Leading Analog and Mixed-Signal Platform
- Analog Devices ADAQ7767-1 μModule DAQ Solution for Rapid Development of Precision Data Acquisition Systems Now Available at Mouser
- Domestic high-precision, high-speed ADC chips are on the rise
- Microcontrollers that combine Hi-Fi, intelligence and USB multi-channel features – ushering in a new era of digital audio
- Using capacitive PGA, Naxin Micro launches high-precision multi-channel 24/16-bit Δ-Σ ADC
- Fully Differential Amplifier Provides High Voltage, Low Noise Signals for Precision Data Acquisition Signal Chain
- Innolux's intelligent steer-by-wire solution makes cars smarter and safer
- 8051 MCU - Parity Check
- How to efficiently balance the sensitivity of tactile sensing interfaces
- What should I do if the servo motor shakes? What causes the servo motor to shake quickly?
- 【Brushless Motor】Analysis of three-phase BLDC motor and sharing of two popular development boards
- Midea Industrial Technology's subsidiaries Clou Electronics and Hekang New Energy jointly appeared at the Munich Battery Energy Storage Exhibition and Solar Energy Exhibition
- Guoxin Sichen | Application of ferroelectric memory PB85RS2MC in power battery management, with a capacity of 2M
- Analysis of common faults of frequency converter
- In a head-on competition with Qualcomm, what kind of cockpit products has Intel come up with?
- Dalian Rongke's all-vanadium liquid flow battery energy storage equipment industrialization project has entered the sprint stage before production
- Allegro MicroSystems Introduces Advanced Magnetic and Inductive Position Sensing Solutions at Electronica 2024
- Car key in the left hand, liveness detection radar in the right hand, UWB is imperative for cars!
- After a decade of rapid development, domestic CIS has entered the market
- Aegis Dagger Battery + Thor EM-i Super Hybrid, Geely New Energy has thrown out two "king bombs"
- A brief discussion on functional safety - fault, error, and failure
- In the smart car 2.0 cycle, these core industry chains are facing major opportunities!
- Rambus Launches Industry's First HBM 4 Controller IP: What Are the Technical Details Behind It?
- The United States and Japan are developing new batteries. CATL faces challenges? How should China's new energy battery industry respond?
- Murata launches high-precision 6-axis inertial sensor for automobiles
- Ford patents pre-charge alarm to help save costs and respond to emergencies
- The topological classification of DCDC can be understood as follows
- HCS12 Series MC9S12DP512MPVE Microcontroller 16Bit Flash 112LQFP
- Annoying technology
- Can a PMOS be connected in series on the communication line as a switch?
- How to achieve positive and negative pulse output using a single battery power supply
- 【McQueen Trial】McQueen's new ultrasonic drive
- Commonly used algorithms for drones - Kalman filter (I)
- 【AT-START-F403A Review】+RTC Test Verification
- 【CH579M-R1】Unboxing and setup environment
- Dial design based on lvgl