I’m creating a commentary system using an FSB that contains around 10000 lines of commentary. I load this FSB using the flags [code:3m847ui5]FMOD_DEFAULT | FMOD_CREATESTREAM | FMOD_NONBLOCKING[/code:3m847ui5]Then when I want to play a line of commentary I use the function setSubSoundSentence on this FSB with the lines of commentary I want to play.
I have found that I am getting some weird static sound at the end of some sounds and that some sounds are missing the first word or two when I play them. E.g. "Here we are at …" will come out as "… we are at …".
This is how I play the sounds:

[code:3m847ui5]m_commentaryFSB->setSubSoundSentence( m_commentaryQueue, m_numberOfSoundsQueued );

m_sysObject->playSound( FMOD_CHANNEL_FREE, m_commentaryFSB, false, &m_commentaryChannel );[/code:3m847ui5]

Has anybody had this issue or knows what is going on?