Service robot is for the purpose of service, so people need a more convenient, more natural and more humanized way to interact with the robot, instead of being satisfied with the complex keyboard and button operation. Human computer interaction based on hearing is an important development direction in this field. At present, the mainstream speech recognition technology is based on statistical pattern. However, due to the complexity of statistical model training algorithm and large amount of computation, it is usually completed by industrial computer, PC or notebook, which undoubtedly limits its application. Embedded voice interaction has become a hot research topic. Compared with the speech recognition system of PC, the embedded speech recognition system has the advantages of small volume, low power consumption, high reliability, small investment, flexible installation and so on, especially suitable for smart home, robot and consumer electronics.

The core processing unit of the module is stm32f103c8t6, a 32-bit processor based on arm Cortex-M3 core of ST company. This module takes the dialogue management unit as the center, realizes the speech recognition function through the hardware unit with ld3320 chip as the core, and adopts the embedded operating system μ C / OS-II to achieve unified task scheduling and peripheral device management. After a large number of experimental data verification, the speech recognition module designed in this paper has the advantages of high real-time, high recognition rate and high stability.

  Speech recognition circuit

Design of embedded speech recognition circuit module based on ARM

Figure 3 is the schematic diagram of speech recognition, which is designed with reference to ld3320 data manual released by icroute. Ld3320 integrates a fast and stable optimization algorithm, which does not need external flash and ram, and does not need user’s prior training and recording to complete speaker independent speech recognition, with high recognition accuracy. In the figure, ld3320 is directly connected with stm32f103c8t6 in parallel mode, both of which are 1K Ω Resistance pull-up, A0 is used to determine whether it is a data segment or an address segment; The control signal, reset signal and interrupt return signal intb are directly connected with stm32f103c8t6, using 10K Ω Resistance pull-up, the auxiliary system works stably; It uses the same external 8 MHz clock as stm32f103c8t6; Led D1 and D2 are used for power on indication after reset; MBS (pin 12) is used as microphone bias, and an RC circuit is connected to ensure that it can output a floating voltage to the microphone.

Leave a Reply

Your email address will not be published. Required fields are marked *