Once all the recording sessions had taken place it was time to prepare the audio and video for editing. The process of aligning the audio to the video is a fairly straight forward one but the fact that the audio existed on various devices and in different formats posed some challenges. These challenges were overcome by adopting creative approaches to monitoring audio and file handling. The audio for each session (bike, snow and ice) was recorded to devices including a Roland portable stereo recorder and an Edirol R4 4 track field recorder. The recorded formats were stereo and B-Format. From these two formats it was necessary to produce 4 individual formats that could be edited alongside the video using Final Cut Pro/Express. These formats were stereo, Binaural, 5.1 and B-format.
The stereo, 5.1 and B-format stems would be derived from the recordings made using the Soundfield microphone and Edirol recorder. The Binaural stem would be taken from the Roland stereo recorder.
Pro Tools
Pro Tools was used to sync all audio and video before any further processes were applied. The video and audio from the camera were imported into the software as and were used as a reference point only. The audio from each device was then imported into the session ready to be synced. Before the B-Format audio could be synchronised a Mid-Side signal had to be extracted from the B-Format stems. The raw B-Format stems were then grouped with the Mid-Side channels and muted (turned off) so that they could be edited but not heard. B-Format audio needs to be processed in one way or another to allow it to be monitored accurately, so in this case the Mid-Side would act as the decode.
This image shows the Mid-Side audio stems and the B-Format stems. As these tracks were 'grouped' together any edits applied to one would be applied to the other. So by using the Mid-Side stems to monitor the audio I was able to move both simultaneously to get them to line up with the video. The audio was lined up using a series of claps at the beginning of the video. This technique is identical to the clapper board used in film. Due to the huge amount of audio and video generated organisation was crucial. At the head of each take a 'slate' would be put down to identify each audio and video sequence. For example, once all audio and video recorders were running I would describe where and when the take was taking place before clapping 5 or 6 times to provide suitable sync points. This proved to be invaluable in the studio when syncing the audio.
The next stage of the process was to 'consolidate' all of the audio files to ensure that they all had an absolute starting point of zero seconds. This would allow the separate audio and video streams to be moved between various software applications for processing and conversion without the risk of them becoming out of sync with each other. Each audio file had to be renamed and appended with the appropriate extension before being exported from the application. For example, if the video was called GoPro 045 then each audio file would carry the same 045 identifier. Again this was another extremely efficient work flow solution to a potentially messy problem.
Adobe Audition
Now that all of the audio had been synchronised I could now use Adobe Audition to clean up any erroneous noises in the recordings. Audition is one of the few audio editing programmes that has a spectral analyser which allows changes in the audio spectrum to be applied through the use of an eraser and pencil etc. This very versatile programme was used to clean up some of the snow and bike audio. The file handling process was the same as described above. The process of cleaning up the audio was the same as discussed in an earlier post
Logic Pro
The next stage of the process was to take all of the clean and consolidated audio into Logic Pro for the final processing stage. The reason that this third piece of audio software was required was that it offered additional functionality not available in my preferred software application Pro Tools. Although similar to some degree I had to quickly grasp the work flow requirements of Logic Audio. This meant spending several hours acquainting myself with the menu structure and additional functionality that this package offered.
Logic was used to produce the 5.1 stems from the B-Format recordings as well as carrying out some equalisation, post Binaural processing and stereo to 5.1 up mixing. The fact that the audio was moving between various applications brought about some other unforeseen problems. These mainly related to file naming conventions and behaviors that were specific to the applications themselves. These problems and how they were addressed will be discussed in the 'Problem with standards post'.
The B-Format to 5.1 conversion was carried out using some free ware plugins produced by Danielle Courville . This suite of plug-ins is a fantastic resource for anyone working with B-Format content. The plug-in that I had initially identified to carry out this task had a license which subsequently expired and therefore forced me to use these instead. As it turned out this proved to be a happy accident as some of the other plug-ins from the suite will be used further down the line for replay etc. The plug-in used to create the 5.1 stems was the B2G. The plug-in took a little time to get used to in terms of the various output options available. Again some of the problems thrown up by these options will be discussed in the 'Problem with standards post'
B2G Plug-In
Ready to Edit
With all of the audio streams now processed I can now begin the task of editing the video and multiple audio streams in Final Cut Express.
No comments:
Post a Comment