An Adaptive Multimodal Voice Communication Method

Abstract

This invention relates to an adaptive multi-mode voice communication method, employing both CMX638 encoding / decoding and PCM encoding / decoding modes. In wired transmission, the voice boxes use CMX638 encoding to generate digitized voice, which is then directly transmitted via the wired connection. At the receiving end, the voice is then decoded using CMX638 to reconstruct the original voice. In wireless radio transmission, the voice boxes use PCM encoding for transmission, which is then converted to high-quality voice at the radio interface before being transmitted wirelessly. This invention effectively resolves the contradiction between wired transmission and analog radio transmission. Furthermore, adaptive decoding of both modes is implemented at the voice box receiving end, achieving seamless switching between the two modes.

Description

Technical Field

[0001] This invention belongs to the field of vehicle-to-vehicle voice communication technology, specifically relating to an adaptive multi-mode voice communication method, and more particularly to a voice communication method based on both CMX638 and PCM encoding / decoding modes. Background Technology

[0002] In harsh environments, there are two main methods for long-distance vehicle-to-vehicle communication: wired transmission and radio wireless transmission. The transmitted content is divided into command data and voice. Due to the long transmission distance, the speed of both methods is significantly limited. In voice transmission, existing solutions primarily use a single mode of digital voice transmission. That is, voice is transmitted digitally in in-vehicle and wired transmissions, while it is transmitted analogally in radio wireless transmissions. Because wired transmission has a long distance and low data transmission speed, it limits the digital-to-analog conversion rate, affecting call quality. Meanwhile, radio wireless transmission involves an additional digital-to-analog conversion, requiring the input analog signal to be as clear and high-quality as possible. This necessitates a high digital-to-analog conversion rate for in-vehicle communication, which contradicts the low bit rate of previous wired voice transmission methods. Summary of the Invention

[0003] This invention provides an adaptive multi-mode voice communication method to solve the following technical problems: 1) In-vehicle voice modes are limited, and voice call quality cannot be selected; 2) Long-distance wired transmission between vehicles has low speed, which limits voice call quality; 3) Inter-vehicle radio voice transmission involves multiple analog-to-digital conversions, resulting in significant voice loss. When the in-vehicle sampling rate is low, the received voice has high distortion, which is not conducive to effective voice transmission; 4) The contradiction between the low speed of wired data transmission and the high sound quality requirements of radio wireless voice.

[0004] To address the above technical problems, this invention provides an adaptive multi-mode voice communication method, characterized by employing both CMX638 encoding / decoding and PCM encoding / decoding modes for voice communication; in wired transmission, the voice boxes use CMX638 encoding to generate digitized voice which is directly transmitted via wired connection, and then decoded back to the original voice at the receiving end using CMX638; in wireless transmission via radio, the voice boxes use PCM encoding for transmission, which is then converted to high-quality voice at the radio interface before being transmitted wirelessly via radio.

[0005] Preferably, the CMX638 codec and PCM codec modes can be selected and switched via buttons on the in-vehicle voice box.

[0006] Preferably, when wired voice transmission is required, the wired button is selected on the voice box panel, and the voice boxes in the vehicle use the CMX638 codec mode; when using a radio for voice transmission, the radio wireless transmission button is selected on the voice box panel, and the voice boxes in the vehicle use the PCM codec mode.

[0007] Preferably, the CMX638 encoding is at a rate of 2.4Kbps, and the data format is to generate 6 bytes of digital voice code every 20ms. The voice code is transmitted to the controller in the voice box via the C-BUS bus for packaging and processing, and finally transmitted between voice boxes via UART.

[0008] Preferably, the PCM encoding mode is 64Kbps, the data format is 1600 bytes of digital voice code generated every 200ms, and the voice data is packaged and processed by the controller and transmitted between voice boxes via UART.

[0009] Preferably, when using CMX638 mode encoding, the encoder CMX638 will generate an interrupt request to the controller to read the encoded data every 60ms. After obtaining the encoded data, the controller will frame the data into a packet after every 3 interrupts and finally send it out through the serial port.

[0010] Preferably, the framed data packets in CMX638 mode mainly consist of a mode field, an address field, a command field, a parameter field, and a check field. At the receiving end, the mode field is mainly used to distinguish the encoding and decoding mode, and also to distinguish whether the communication method is wired or wireless. The address field mainly indicates the address of the voice box encoding and decoding device. The command field and parameter field mainly refer to the type and content of the data packet. The check field is for the reliability of data packet transmission.

[0011] Preferably, the frame data packet format of the PCM encoding mode is the same as that of the CMX638 mode, consisting of a mode field, an address field, an instruction field, a parameter word field, and a check field. The only difference is that the mode word that distinguishes the encoding mode is changed to DC92, and the data size of the parameter word field is changed to 1600 bytes.

[0012] Preferably, the receiving end needs to first determine the format of the voice data packet: the receiving end is equipped with an adaptive decoding module, which identifies the format of the voice data through the mode field in the data packet. That is, when it is determined that valid voice data has been received in the receiving buffer, the mode word in the mode field is extracted first, and the CMX638 voice packet or PCM voice packet is identified based on the mode word. Then, the analog voice is decoded and output according to the corresponding data packet format.

[0013] Beneficial effects: In wired transmission, this invention uses CMX638 encoding to generate digitized voice between voice boxes for direct wired transmission, and then decodes the voice at the receiving end using CMX638 to restore the original voice. In analog radio transmission, PCM encoding is used between voice boxes for transmission, which is then converted to high-quality voice at the radio interface before being transmitted analogly, effectively resolving the contradiction between wired and analog radio transmission. Simultaneously, adaptive decoding of both modes is implemented at the voice box receiving end, achieving seamless switching between the two modes. Attached Figure Description

[0014] Figure 1 Multi-mode voice box composition block diagram

[0015] Figure 2 CMX638 Frame Timing Relationships

[0016] Figure 3 PCM encoding timing relationship

[0017] Figure 4 This is the flowchart for adaptive decoding. Detailed Implementation

[0018] To make the objectives, contents, and advantages of the present invention clearer, the specific embodiments of the present invention will be described in further detail below.

[0019] Voice communication is an indispensable part of vehicle communication systems. However, existing single-mode wired transmission and radio analog voice signal transmission systems present a trade-off between digital bandwidth and analog sound quality.

[0020] The present invention proposes an adaptive multi-mode voice communication method, characterized by: employing both CMX638 encoding / decoding and PCM encoding / decoding modes for voice communication; in wired transmission, the voice boxes use CMX638 encoding to generate digitized voice which is directly transmitted via wired connection, and then decoded by CMX638 at the receiving end to restore the voice; in radio wireless transmission, the voice boxes use PCM encoding for transmission, which is converted to high-quality voice at the radio interface before being transmitted wirelessly via the radio.

[0021] This invention provides two encoding / decoding modes for in-vehicle voice communication: CMX638 and PCM. These modes can be switched using buttons on the in-vehicle voice box. Since inter-vehicle communication primarily involves wired transmission and wireless radio transmission, this invention selects two encoding / decoding modes: one for wired transmission and one for wireless radio transmission. When wired voice transmission is required, the wired button is selected on the voice box panel. The in-vehicle voice boxes use the CMX638 encoding / decoding mode, with a CMX638 encoding rate of 2.4Kbps and a data format... The system generates 6 bytes of digital voice code every 20ms. The voice code is transmitted to the controller in the voice box via the C-BUS bus for packaging and processing, and finally transmitted between the voice boxes via UART. When using a radio for voice transmission, the radio wireless transmission button needs to be selected on the voice box panel. At this time, the voice boxes in the vehicle use PCM encoding and decoding mode. The PCM encoding mode has a rate of 64Kbps, and the data format is to generate 1600 bytes of digital voice code every 200ms. The voice data is packaged and processed by the controller and then transmitted between the voice boxes via UART.

[0022] Figure 2 The timing relationship for voice data framing in CMX638 mode is as follows: When encoding in CMX638 mode, 6 bytes of voice data are generated every 20ms. When the controller receives the encoded data, it frames and packages this voice data. The framing relationship is as follows: Figure 2 As shown, the CMX638 encoder generates an interrupt request to the controller to read the encoded data every 60ms. After receiving the encoded data, the controller frames the data into a packet every three interrupts and finally sends it out via the serial port. Table 1 shows the framed data packet format in CMX638 mode, which mainly consists of a mode field, address field, command field, parameter field, and check field. At the receiving end, the mode field is mainly used to distinguish the encoding and decoding mode and also to distinguish whether the communication method is wired or wireless. The address field mainly indicates the address of the voice box encoding and decoding device to prevent communication conflicts. The command field and parameter field mainly refer to the type and content of the data packet. The check field is for the reliability of data packet transmission.

[0023] Table 1: Framed data packet format in CMX638 mode

[0024]

[0025] Figure 3The timing relationship for framing voice data in PCM mode is explained. Since PCM encoding is 64Kbps (8K sampling rate, 8-bit encoding width) and is continuous, the data volume is relatively large. Therefore, this invention uses 1600 bytes per 200ms framed data packet, which is then sent via UART after being packaged. Table 2 shows the framed data packet format for PCM encoding mode. Similar to CMX638 mode, it consists of a mode field, address field, instruction field, parameter word field, and check field. The only difference is that the mode word distinguishing the encoding mode is changed to DC92, and the data size of the parameter word field is changed to 1600 bytes.

[0026] Table 2: Framed data packet format of PCM encoding mode

[0027]

[0028] In the voice box, to distinguish between wired and wireless modes, selection can be made via buttons to switch voice encoding modes and output encoded data. At the receiving end, to differentiate the voice data formats in different modes for correct decoding, the format of the voice data packets must first be determined. The voice data receiving module requires an adaptive decoding module. This module is responsible for identifying the voice data format through the mode field in the data packets. Specifically, upon determining that valid voice data has been received in the receive buffer, it first extracts the mode word from the mode field, identifies the CMX638 or PCM voice packet based on the mode word, and then decodes and outputs analog voice according to the corresponding data packet format. Figure 4 The adaptive decoding flowchart shows that after the voice box completes power-on initialization, the adaptive decoding module first determines whether the receive buffer has received data. If data is received, it determines whether it is local data according to the address field protocol in the data packet. If it is confirmed to be local data, it parses the mode word. If the mode word is an EBA0 frame header, it starts CMX638 decoding to output analog voice; if the mode word is a DC92 frame header, it starts PCM decoding to output analog voice; if neither is true, it is discarded.

[0029] This invention employs a dual-mode method with high and low sampling rates within the in-vehicle voice box. Panel buttons differentiate between in-vehicle conversations, wired connections between vehicles, and in-vehicle radio broadcasts. For wired connections, a low bitrate voice transmission mode is used due to the lower transmission rate. Since the wired connection between the two vehicles is direct and voice transmission between wired connections is digital, the wired voice transmission is acceptable. For in-vehicle and radio voice transmissions, a high bitrate voice acquisition mode is used. This ensures that voice intelligibility is maintained even after multiple samplings, effectively resolving the contradiction between the low data transmission rate of wired connections and the high audio quality requirements of radio wireless voice broadcasts.

[0030] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An adaptive multi-mode voice communication method, characterized in that: Voice communication employs both CMX638 and PCM codec modes. In wired transmission, CMX638 encoding is used between the voice boxes at a rate of 2.4Kbps, with a data format of 6 bytes of digital voice code generated every 20ms. The generated digitized voice is directly transmitted via wired connection and then decoded at the receiving end using CMX638 to restore the original voice. In wireless transmission via radio, PCM encoding is used between the voice boxes at a rate of 64Kbps, with a data format of 1600 bytes of digital voice code generated every 200ms. This is converted to high-quality voice at the radio interface before being transmitted wirelessly. The CMX638 and PCM codec modes are selected and switched using buttons on the in-vehicle voice box.

2. The adaptive multi-mode voice communication method according to claim 1, characterized in that: When wired voice transmission is required, select the wired button on the voice box panel. The voice boxes in the vehicle use the CMX638 codec mode. When using the radio for voice transmission, select the radio wireless transmission button on the voice box panel. In this case, the voice boxes in the vehicle use the PCM codec mode.

3. The adaptive multi-mode voice communication method according to claim 1, characterized in that: When using the CMX638 encoding mode, the voice code is transmitted to the controller in the voice box via the C-BUS bus for packaging and processing, and finally transmitted between voice boxes via UART.

4. The adaptive multi-mode voice communication method according to claim 1, characterized in that: When using PCM encoding mode, voice data is packaged and processed by the controller and then transmitted between voice boxes via UART.

5. The adaptive multi-mode voice communication method according to claim 3, characterized in that: When using CMX638 mode encoding, the encoder CMX638 will generate an interrupt request to the controller to read the encoded data every 60ms. After obtaining the encoded data, the controller will frame the data into a packet after every 3 interrupts and finally send it out through the serial port.

6. The adaptive multi-mode voice communication method according to claim 5, characterized in that: In CMX638 mode, framed data packets mainly consist of a mode field, an address field, a command field, a parameter field, and a check field. At the receiving end, the mode field is mainly used to distinguish the encoding / decoding mode and also to distinguish whether the communication method is wired or wireless. The address field mainly indicates the address of the voice box encoding / decoding device. The command field and parameter field mainly refer to the type and content of the data packet. The check field is for the reliability of data packet transmission.

7. The adaptive multi-mode voice communication method according to claim 4, characterized in that: The frame data packet format of PCM encoding mode is the same as that of CMX638 mode, consisting of a mode field, address field, instruction field, parameter word field, and check field. The only difference is that the mode word that distinguishes the encoding mode is changed to DC92, and the data size of the parameter word field is changed to 1600 bytes.

8. An adaptive multi-mode voice communication method according to any one of claims 1-7, characterized in that: At the receiving end, it is necessary to first determine the format of the voice data packet: The receiving end is equipped with an adaptive decoding module, which identifies the format of the voice data through the mode field in the data packet. That is, when it is determined that valid voice data has been received in the receiving buffer, the mode word in the mode field is extracted first, and the CMX638 voice packet or PCM voice packet is identified based on the mode word. Then, the analog voice is decoded and output according to the corresponding data packet format.

Images