Shahzad wrote:

so this is requirement that the message will not go on the speaker.... and to stop this that menas we need to cut off speaker from ISD.

So, you need a muting ability for the ISD signal, Shahzad?

You could use a 1/4 CD4066 to switch-off ISD signal, whenever needed: Connect according control input to some free port pin of '51 micro. And whenever this port pin goes low, analog switch of CD4066 is opened and ISD signal is muted.

Have a look at this schematic:



CD4066 provides an off-isolation of about 50dB at 1kHz in this configuration. Added distortion is less than 0.5%. But, of course, you could also use a more sophisticated analog switch, providing much better off-isolation and lower distortion, if needed. E.g. by the use of 74HC4066 and 'T-switch' configuration distortion can be reduced to much less than 0.1% and off-isolation becomes so high, that you can hear the electrons breathing...


Two comments:

1. Control line of CD4066 is connected to port pin of '51 micro while using pull-up resistor of 10k. This resistor is optional, but helps to improve signal integrity, when processor emits logical high state. As you surely know, quasibidirectional port pin configuration of standard '51 uses an array of active pull-ups, which provides a rather high impedant path to Vcc. This impedance is much higher than typical 74HCMOS output impedance and can make trouble, when high frequency interference is capacitively coupled into your system. Then, output impedance is not low enough for effectively shunting interference to Vcc. In this case use of pull-ups is of benefit, especially if CD4066 is located 'far' away from micro.

2. Next to CD4066 you will find two 10k resistors connected as voltage divider. Voltage divider is needed to fabrictae Vcc/2 at left side of 100nF capacitor. This helps to prevent 'pop' signal, when CD4066 is turned-on.
Without voltage divider 100nF capacitor would need some time to be charged to DC output voltage of ISD1212. The result would be a 'pop' signal. Another advantage of voltage divider is, that LM386 is not overdriven, when 100nF has to be charged (see last reply).
By the way, voltage divider also helps, even if CD4066 is not inserted at all. Why? Because ISD1212 is mostly in standby mode, when not emitting its message. And in this condition output of ISD1212 is switched-off, means output impedance is very high ohmic, like a disabled tristate output. So, whenever ISD1212 starts its message, aprupt develop of DC voltage at output of ISD1212 could cause an overdrive of LM386 or fabrication of unwanted 'pop' signal.

For best performance, fabricated DC voltage on left side of 100nF capacitor (by the help of voltage divider) should equal DC output voltage of ISD1212. I assumed this to be Vcc/2, which is mostly the case for bridge driver outputs. But if you observe a different DC voltage at output of ISD1212 you should adapt the voltage divider.

Good luck,
Kai