Hi,
Play around with lame options a bit. First of all '-s 8000' is wrong, it
should be '-s 8'. I don't know what you do when recording both directions, but
having two PCM files, you need to mix them, so they don't clip.
I would suggest using Audacity for this audio work.
Thanx,
Jaap
TORKHANI Wajdi wrote:
Thank you so much
After 4 weeks I can finally to hear the voice,
But now I have another problem!
If i record only one direction the sound run very quickly and if record
both direction i have a very very very bad quality of voice !!!!
1-I record the payload in a binary file:
Code:
void rawfile(unsigned char * payload,FILE *f){
fwrite(payload, 10,1,f); //10 :because audio data is packed into 80
bits (10 bytes)
}
2- I decode the binary file by using voiceage G729, which gives me a
file : "16-bit mono PCM speech data sampled at 8000 Hz"
3- convert PCM file into wav by using Lame with:
lame --decode -x -r -s 8000 -m m -b 16 file.pcm file.wav
Is it correct ?!
----- Original Message ----- From: "Jaap Keuter" <jaap.keuter@xxxxxxxxx>
To: "Developer support list for Wireshark" <wireshark-dev@xxxxxxxxxxxxx>
Sent: Saturday, March 15, 2008 8:46 PM
Subject: Re: [Wireshark-dev] RTP raw file
Hi,
Have a look at rtp_analysis.c
The idea is to 'tap' the RTP packets and write out the payloads into a
file.
That gives you the stream as seen on the network, actual voip
applications use
a jitterbuffer to recover from sequence errors and changing network
delay.
Thanx,
Jaap
TORKHANI Wajdi wrote:
Hello,
Im a final year student of engineering. Im doing a project on VOIP
(G.729)
I must create a voip sniffer (to capture communication VOIP on the LAN)
and then to convert them into audio format.
I succeeded in preparing a sniffer in C++ (by using the library
winpcap) to:
1- capture network traffic
2- Filtre UDP trafic
3- Read ethernet,ip,udp and RTP header.
But I do not know how to create a raw file seems the output file created
by Ethereal
(Statistics->RTP->STeam Analysis->save payload-> .raw )
Please please help me.
Thank you so much.
