Voice Codecs

G711u (PCMU)

• 64 kbps data rate
• 10 ms sample size
• Default datagram 2 samples = 20 ms
• Default 1.5 ms packetization delay
• Default 40 ms jitter buffer delay (based on 2 datagrams)
• Theoretical maximum MOS 4.4

G711a (PCMA)

• 64 kbps data rate
• 10 ms sample size
• Default datagram 2 samples = 20 ms
• Default 1.5 ms packetization delay
• Default 40 ms jitter buffer delay (based on 2 datagrams)
• Theoretical maximum MOS 4.4

G729a

• 8 kpbs data rate
• 10 ms frame size
• Default datagram 2 frames = 20ms
• Default 15.0 ms packetization delay
• Default 40 ms jitter buffer delay (based on 2 datagrams)
• Theoretical maximum MOS 4.07

G723.1

• 6.3 kbps
• 30 ms frame size
• Default datagram 1 frame = 30 ms
• Default 37.5 ms packetization delay
• Default 60 ms jitter buffer delay (based on 2 datagrams)
• Theoretical maximum MOS 3.87
• g723.1 can also be 5.3 kbps with a maximum MOS 3.69

G722 Wideband

• 64kbps, 56kbps, 48kbps
• Signal bandwidth 7khz

Note: G711 is a sample based codec G723 and G729 are frame based codecs

Sample size determination by looking at RTP payload

Sample and Frame based codecs

• Count the number of bytes in the RTP payload
• Convert the number of bytes into bits
• Multiply the number of bits by 1/(Band Width)

Example: 160 bytes * 8 bits/byte * sec/(64 103 bits) = 20 * 10*-3 = 20 ms