2024年3月11日发(作者：寸雅辰)
The introduce of sound playback chip ISD25120
1 ISD 2500 series chip Profile
ISD's 2500 series chips, according to recording time 60 seconds, 75 seconds, 90
seconds and 120 seconds into ISD2560, 2575,2590 and 25120 four models. ISD
device with OVF (overflow) client for cascading multiple devices.
ISD2500 series EEPROM with capacity for 480K, up to sub-paragraph 600. Four
different types of recording time depends on different input sampling rate to achieve,
and they were: 8.0,6.4,5.3,4.0 kHz.
DIP package 28-pin devices, the pin function as follows:
1 ~ 7 A0/M0 ~ A6/M6 address / mode selection
8 ~ 10 A7 ~ A9 enter an address line
11 AUX IN input
12,13 VSSD, VSSA and analog to digital
14,15 SP +, SP-speaker output
16 VCCA power analog signal positive
17,18 MIC, MIC REF
Microphone input and the input reference terminal
19 AGC AGC
20,21 ANA IN, ANA OUT
Analog signal input and output
Spillage 22 OUF
23 CE chip select (low to allow the chip work)
24 PD-chip low-power state control
25 EOM an end audio playback signal output
26 XCLK external clock
27 P / R recorded / release control to choose
28 VCCD cathode power supply digital signal
2 Working Principle
2500 Series 10 input address A0 ~ A9, the site recorded up to 1024, address space
is 0 ~ 1023. Its distribution is: Address 0 ~ 299 as a sub, and did not address the use
of 600 ~ 767, 768 ~ 1023 for the address mode selection (ie, A8, A9 are high).
Address Line 2500 series has two purposes, first as a mode of control, the second
sub-recording sound as the start segment address. When the highest address bit (MSB)
A8, A9 are high (ie address 768 ~ 1023), the address A0 ~ A6-side as a working
model on the choice of client M0 ~ M6, corresponding to seven kinds of work. When
the A8, A9 or any one for the low low (that is, address 0 ~ 599), as long as recorded
in the sub / playback pre-operation (less than 300ns) to the address A0 ~ A9
assignment operation from the Address to start.
2500 series of voice chips will be divided into 480K of EEPROM information in
paragraph 600, 800 bytes each. Unit as a whole, addressing and control information to
each paragraph should be assigned an address for the external control rather than for
each byte addressing, or at least address port 19. This greatly reduces the information
needed to retrieve the address line. Of the speech signal can be a long cross-section of
a number of information recording, storage information from the section of the
internal constraints, and information within the site will automatically increase the
segment. Each voice in the automatic increase in the tail section of an end mark EOM,
when the combination of playing, by detecting EOM to control the end of the voice
segment and the beginning of the next section.
Information section of each audio playback time equal to the total time divided by
600. ISD2560, such as the total time for the 60s, the information section of each audio
playback time of 100ms; ISD25120 the total time for the 120s, each time for
paragraph information 200ms. Therefore the length of time can be used as a segment
address, through the time-parallel single-chip timer to the address above mapping
information in order to get the start address of each recording. In this way, on the need
to set up an address counter. General address from 0 to start recording, first of all,
through the CPU assigned to it A0 ~ A9, then start recording ISD SCM, SCM to start
the timer at the same time the beginning of time, every time an information paragraph,
give the address count plus 1. ISD single-chip control when to stop recording at the
same time stop the timer time. At this point the value of address counter is the voice
of the paragraph did not address, plus 1 under the section of speech is the first address,
and EEPROM in its existence, will be playing for the next address information
provided. Through the CPU assigned to the address A0 ~ A9, under the section of
your voice recording. Followed it, you can record voice above all at the same time be
the starting address of the paragraphs. If it is not started from 0 voice above address,
simply assign the initial address A0 ~ A9, with the value of address counter can be a
voice at the end of the address above. Not here at the same time to preserve the voice
segment start address and end address, because it is adjacent to each paragraph, before
paragraph 1 at the end of the address that is the starting address of this paragraph, and
each have a voice at the end of paragraph mark EOM may give a disruption. Use it
when playing and stored in the EEPROM in the initial paragraph of the speech
address in any order combination of voice above all.
ISD25120录放音芯片介绍
1 ISD 2500系列芯片简介
ISD公司的2500系列芯片，按录放时间60秒、75秒、90秒和120秒分成
ISD2560、2575、2590和25120四个型号。ISD器件设有OVF（溢出）端，便于
多个器件级联。
ISD2500系列片内EEPROM容量都为480K，最多能分600段。四个型号的
不同录放时间是靠不同的输入采样率来实现的，它们分别为：8.0、6.4、5.3、4.
0kHz。
DIP器件封装为28脚，各引脚功能如下：
1～7 A0/M0～A6/M6地址/模式选择
8～10 A7～A9输入地址线
11 AUX IN辅助输入
12、13 VSSD、VSSA数字地和模拟地
14、15 SP+、SP-扬声器输出
16 VCCA模拟信号电源正极
17、18 MIC、MIC REF
麦克风输入端和输入参考端
19 AGC自动增益控制
20、21 ANA IN、ANA OUT
模拟信号输入和输出
22 OUF 溢出
23 CE 片选（低电平允许芯片工作）
24 PD 芯片低功耗状态控制
25 EOM 录放音结束信号输出
26 XCLK 外部时钟
27 P/R 录/放控制选择
28 VCCD 数字信号电源正极
2 工作原理
2500系列有10个地址输入端A0～A9，录址能力可达1024位，地址空间
为0～1023。其分配情况是：地址0～299作为分段用，地址600～767未使用，
地址768～1023为工作模式选择（即A8、A9均为高）。2500系列的地址线有
两种用途，一是作为工作模式控制，二是作为分段录放音的起始段地址。当最高
位地址（MSB）A8、A9都为高电平时（即地址768～1023），地址端A0～A6
就作为工作模式选择端M0～M6，对应7种工作模式。当A8、A9任一位为低或
都为低时（即地址0～599），只要在分段录/放音操作前（不少于300ns）给地
址A0～A9赋值，操作就从该地址开始。
2500系列语音芯片将480K的EEPROM分为600个信息段，每段800个字
节。作为一个整体单位进行寻址和控制，应给每个信息段分配一个供外部控制的
地址，而不是对每个字节进行寻址，否则至少需要19个地址端口。这样，大大
减少了信息检索所需要的地址线。对较长的语音信号可以跨越多个信息段进行录
音，不受内部存储信息段的限制，且内部的信息段址会自动增加。在每个语音段
的尾部自动增加一个结束标志EOM，组合放音时，通过检测EOM来控制各语
音段的结束和下一段的开始。
每个信息段的录放音时间等于总时间除以600。如ISD2560的总时间为60s，
则每个信息段的录放音时间为100ms；ISD25120的总时间为120s，则每个信息
段的时间为200ms。因此可以利用该时间长度作为一个段地址，通过单片机定时
器的计时平行地映射信息段的地址，从而得到每段录音的起始地址。这样，就需
要设置一个地址计数器。一般录音从0地址开始，首先通过CPU将它赋给A0～
A9，然后通过单片机控制ISD启动录音，同时启动单片机的定时器开始计时，
每到一个信息段的时间，就给地址计数加1。当单片机停止控制ISD录音时，同
时停止定时器计时。此时地址计数器的值即为该段语音的未地址，加1即为下一
段语音的首地址，并将它存在EEPROM中，为下一将放音提供的地址信息。通
过CPU将该地址赋给A0～A9，即可录制下一段语音。依次下去，即可在录制
完所有语音段的同时得到各段的起始地址。如果不是从0地址开始的语音段，只
需将初始地址赋给A0～A9，加上地址计数器的值，即可得到语音段的末地址。
这里不用同时保存各语音段的起始地址和结束地址，因为各个段是相邻的，前一
段的末地址加1即是本段的起始地址，且每个语音段的结尾均有EOM标志，并
可发出中断。放音时利用它和保存在EEPROM中各语音段的起始地址即可按任
意顺序组合各个语音段。
2024年3月11日发(作者：寸雅辰)
The introduce of sound playback chip ISD25120
1 ISD 2500 series chip Profile
ISD's 2500 series chips, according to recording time 60 seconds, 75 seconds, 90
seconds and 120 seconds into ISD2560, 2575,2590 and 25120 four models. ISD
device with OVF (overflow) client for cascading multiple devices.
ISD2500 series EEPROM with capacity for 480K, up to sub-paragraph 600. Four
different types of recording time depends on different input sampling rate to achieve,
and they were: 8.0,6.4,5.3,4.0 kHz.
DIP package 28-pin devices, the pin function as follows:
1 ~ 7 A0/M0 ~ A6/M6 address / mode selection
8 ~ 10 A7 ~ A9 enter an address line
11 AUX IN input
12,13 VSSD, VSSA and analog to digital
14,15 SP +, SP-speaker output
16 VCCA power analog signal positive
17,18 MIC, MIC REF
Microphone input and the input reference terminal
19 AGC AGC
20,21 ANA IN, ANA OUT
Analog signal input and output
Spillage 22 OUF
23 CE chip select (low to allow the chip work)
24 PD-chip low-power state control
25 EOM an end audio playback signal output
26 XCLK external clock
27 P / R recorded / release control to choose
28 VCCD cathode power supply digital signal
2 Working Principle
2500 Series 10 input address A0 ~ A9, the site recorded up to 1024, address space
is 0 ~ 1023. Its distribution is: Address 0 ~ 299 as a sub, and did not address the use
of 600 ~ 767, 768 ~ 1023 for the address mode selection (ie, A8, A9 are high).
Address Line 2500 series has two purposes, first as a mode of control, the second
sub-recording sound as the start segment address. When the highest address bit (MSB)
A8, A9 are high (ie address 768 ~ 1023), the address A0 ~ A6-side as a working
model on the choice of client M0 ~ M6, corresponding to seven kinds of work. When
the A8, A9 or any one for the low low (that is, address 0 ~ 599), as long as recorded
in the sub / playback pre-operation (less than 300ns) to the address A0 ~ A9
assignment operation from the Address to start.
2500 series of voice chips will be divided into 480K of EEPROM information in
paragraph 600, 800 bytes each. Unit as a whole, addressing and control information to
each paragraph should be assigned an address for the external control rather than for
each byte addressing, or at least address port 19. This greatly reduces the information
needed to retrieve the address line. Of the speech signal can be a long cross-section of
a number of information recording, storage information from the section of the
internal constraints, and information within the site will automatically increase the
segment. Each voice in the automatic increase in the tail section of an end mark EOM,
when the combination of playing, by detecting EOM to control the end of the voice
segment and the beginning of the next section.
Information section of each audio playback time equal to the total time divided by
600. ISD2560, such as the total time for the 60s, the information section of each audio
playback time of 100ms; ISD25120 the total time for the 120s, each time for
paragraph information 200ms. Therefore the length of time can be used as a segment
address, through the time-parallel single-chip timer to the address above mapping
information in order to get the start address of each recording. In this way, on the need
to set up an address counter. General address from 0 to start recording, first of all,
through the CPU assigned to it A0 ~ A9, then start recording ISD SCM, SCM to start
the timer at the same time the beginning of time, every time an information paragraph,
give the address count plus 1. ISD single-chip control when to stop recording at the
same time stop the timer time. At this point the value of address counter is the voice
of the paragraph did not address, plus 1 under the section of speech is the first address,
and EEPROM in its existence, will be playing for the next address information
provided. Through the CPU assigned to the address A0 ~ A9, under the section of
your voice recording. Followed it, you can record voice above all at the same time be
the starting address of the paragraphs. If it is not started from 0 voice above address,
simply assign the initial address A0 ~ A9, with the value of address counter can be a
voice at the end of the address above. Not here at the same time to preserve the voice
segment start address and end address, because it is adjacent to each paragraph, before
paragraph 1 at the end of the address that is the starting address of this paragraph, and
each have a voice at the end of paragraph mark EOM may give a disruption. Use it
when playing and stored in the EEPROM in the initial paragraph of the speech
address in any order combination of voice above all.
ISD25120录放音芯片介绍
1 ISD 2500系列芯片简介
ISD公司的2500系列芯片，按录放时间60秒、75秒、90秒和120秒分成
ISD2560、2575、2590和25120四个型号。ISD器件设有OVF（溢出）端，便于
多个器件级联。
ISD2500系列片内EEPROM容量都为480K，最多能分600段。四个型号的
不同录放时间是靠不同的输入采样率来实现的，它们分别为：8.0、6.4、5.3、4.
0kHz。
DIP器件封装为28脚，各引脚功能如下：
1～7 A0/M0～A6/M6地址/模式选择
8～10 A7～A9输入地址线
11 AUX IN辅助输入
12、13 VSSD、VSSA数字地和模拟地
14、15 SP+、SP-扬声器输出
16 VCCA模拟信号电源正极
17、18 MIC、MIC REF
麦克风输入端和输入参考端
19 AGC自动增益控制
20、21 ANA IN、ANA OUT
模拟信号输入和输出
22 OUF 溢出
23 CE 片选（低电平允许芯片工作）
24 PD 芯片低功耗状态控制
25 EOM 录放音结束信号输出
26 XCLK 外部时钟
27 P/R 录/放控制选择
28 VCCD 数字信号电源正极
2 工作原理
2500系列有10个地址输入端A0～A9，录址能力可达1024位，地址空间
为0～1023。其分配情况是：地址0～299作为分段用，地址600～767未使用，
地址768～1023为工作模式选择（即A8、A9均为高）。2500系列的地址线有
两种用途，一是作为工作模式控制，二是作为分段录放音的起始段地址。当最高
位地址（MSB）A8、A9都为高电平时（即地址768～1023），地址端A0～A6
就作为工作模式选择端M0～M6，对应7种工作模式。当A8、A9任一位为低或
都为低时（即地址0～599），只要在分段录/放音操作前（不少于300ns）给地
址A0～A9赋值，操作就从该地址开始。
2500系列语音芯片将480K的EEPROM分为600个信息段，每段800个字
节。作为一个整体单位进行寻址和控制，应给每个信息段分配一个供外部控制的
地址，而不是对每个字节进行寻址，否则至少需要19个地址端口。这样，大大
减少了信息检索所需要的地址线。对较长的语音信号可以跨越多个信息段进行录
音，不受内部存储信息段的限制，且内部的信息段址会自动增加。在每个语音段
的尾部自动增加一个结束标志EOM，组合放音时，通过检测EOM来控制各语
音段的结束和下一段的开始。
每个信息段的录放音时间等于总时间除以600。如ISD2560的总时间为60s，
则每个信息段的录放音时间为100ms；ISD25120的总时间为120s，则每个信息
段的时间为200ms。因此可以利用该时间长度作为一个段地址，通过单片机定时
器的计时平行地映射信息段的地址，从而得到每段录音的起始地址。这样，就需
要设置一个地址计数器。一般录音从0地址开始，首先通过CPU将它赋给A0～
A9，然后通过单片机控制ISD启动录音，同时启动单片机的定时器开始计时，
每到一个信息段的时间，就给地址计数加1。当单片机停止控制ISD录音时，同
时停止定时器计时。此时地址计数器的值即为该段语音的未地址，加1即为下一
段语音的首地址，并将它存在EEPROM中，为下一将放音提供的地址信息。通
过CPU将该地址赋给A0～A9，即可录制下一段语音。依次下去，即可在录制
完所有语音段的同时得到各段的起始地址。如果不是从0地址开始的语音段，只
需将初始地址赋给A0～A9，加上地址计数器的值，即可得到语音段的末地址。
这里不用同时保存各语音段的起始地址和结束地址，因为各个段是相邻的，前一
段的末地址加1即是本段的起始地址，且每个语音段的结尾均有EOM标志，并
可发出中断。放音时利用它和保存在EEPROM中各语音段的起始地址即可按任
意顺序组合各个语音段。