Construct multi-audio video streaming ism, ismc files

Question

Hi, we experienced a critical problem on combining multi-audio videos in Expression Encoder where we previously posted a question here:
http://social.expression.microsoft.com/Forums/en-US/encoder/thread/80ab88ac-9c86-46a0-9edf-d6f7abee92a9

and similar question was asked by others before like:
http://smf.codeplex.com/discussions/216664

We figured out a manual method to do so, but another critical problem is noticed. The method is described as follows:

Combining audio’s and video’s ism files

1. Copy <audio> ... </audio> tag from audio’s ism file to the position after </video> tag in video’s ism file
2. Search and Change the flag HasAudio to True in video’s ism file

Combine audio’s and video’s ismc files

1. Copy <StreamIndex> ... </StreamIndex> tag from audio’s ismc file to the position after </StreamIndex> tag in video’s ismc file
2. Search the audio tag in video’s ismc file in and change the Name from “audio” to the language name of the audio track, e.g. “english”

A screenshot of manual changes is shown below:

The result is that, when one audio track is selected (e.g. English), the other audio track (e.g. Chinese) will appear randomly too. This happened every time and cannot be solved. Besides, the video keeps jumping and skipping. However, all these do not appear for those with only single audio track.

Please kindly help to suggest multi-audio ism and ismc files conversion methods with Expression Encoder or if there's another method to avoid the above problems for manually combing multi-audio video files could be suggested. Deeply appreciated for your help.