i am finishing a project and want to tweak my sound levels myself rather than go to a dub. i need to increase the level for final output on dvd.
- the audio is excellent (recorded at -18db). all i need to do is raise the level.
- the audio mono speech and there are no music and effects.
- all the audio was recorded in the same place in a studio and is within a similar range.
i want to raise the level to -6db (average)
1. if i modify the sound within a quicktime on my fcp timeline by opening an exernal editor (soundtrack or sound studio) and normalize (at -6db) the change is overwritten on my source file which strikes me as possibly dangerous, but easy.
2. is it better to raise the level on the audio within fcp?
3. or is better to output final wavs/aifs and raise the level on those files in soundtrack or sound studio?
what is good practice?
'ppreciate input from a sound pro...
this is a video project which i'm cutting on final cut pro. the project is 90 minutes long, and the source footage is captured by tape in 40 minute segments. i have 10 hours of footage from which i have cut my final programme.
Try nesting the sequence then just raise the overall level in FCP before output.
do you not think it better to normailise the file to increase the dynamic range without clipping? raising the level in fcp just increases the volume, and does not increase the dynamic range.
any sound pros out there who can advise?
Dynamic range is the difference between the lowest level of audio and the highest level of audio.
A true "Normalize" should have no effect on the dynamic RANGE (it just makes sure that the highest levels in a clip are reaching the maximum set up in the Normalize filter or device),
Adjusting the overall level ("master volume") will have no effect on dynamic range either.
ITOH, a "Compressor" is designed to DECREASE the dynamic range, and an "Expander" is designed to INCREASE the dynamic range.
thanks for that. still a little perplexed - although you were very clear. when i normalise a sound clip the sound wave increases in size but when i raise the level the sound wave remains the same - ie just the level of db is raised. what is the difference? what is happening to the sound wave, and why does it increase in graphical size?
There are different versions of filters that might be "called" normalize, but are, in fact, compressors.
And, many of the various terms are used interchangeably by different operators and designers.
Regardless of the terminology, just use the filter that does what you want it to do.
When you apply the "normalize" filter, it creates a new copy of the file with the gain increase applied.
When you increase the level of a track using the software's faders, you are increasing the gain in much the same way except that the original audio file is not replaced (ie. the level increase is being applied to the same audio during playback, but the audio file itself is unchanged). This is called a non-destructive opperation, and has the advantage of easily being tweaked or modified. Such fader manipulations are generally processed in real-time by the software, making them much more flexible than rendered effects. Just bear in mind that you will have to re-export the program in order to render in the fader adjustments.
Try not to rely too much on the thumbnail view of the audio waveform shown on the track in the timeline. When it comes to the final mixing pass, the VU meters are a much more accurate tool for judging the audio level. And even the VU meters are a poor substitute for a decent set of near-field monitors and a propperly tuned monitoring environment.
stephen, thanks. that's great. i'm well aware of the vu/ppm issues etc but i still would like to know which you would recommend:
my sound was recorded beautifully but i need to increase the level for output.
1. would you increase the level by raising the db on the timeline - grabbing the 'rubber band' and increasing the level,
2. or would you normalise it and set the level within the external software? (ignoring any additional work needed with the creation of new files)
another way of asking the question is if i was to take my edit to dubbing mixer in a sound studio and ask him to increase the levels etc what method would he use?
sounds pedantic but i don't have proper broad spec monitoring yet so am i'm trying to get an idea of what is regarded as good practice. the impression i'm getting at the moment is that what sounds right is right. when i'm grading pictures this rule is also true but there are broadcast safe specs and methods which apply and i work within. i'm trying to find out as much as i can about simple sound mixing so i can get decent results in the same way.
Although the questions you ask are simple enough, the answers can get extremely complex.
Here's an example of a typical chain of events from non-linear picture editorial to final mix:
- The programme's picture edit is locked.
- The picture is output to a work tape (eg. BetaSP, DVcam) or video clip. This work tape or clip will include the 2-beep and timecode burn-in.
- The audio content of the picture timeline is exported as an audio OMF (a media data interchange format available in most professional editing software) with the appropriate amount of "handle" on the regions (eg. 5 seconds).
- The OMF files and work tape/clip are handed to the sound editor to import into (typically) a ProTools session for audio post-production
- Depending on the complexity of the project, a number of specialised sound editors will begin working on the dialogue, effects, and music.
- Spotting notes are prepared for any foley or ADR required.
- The foley and ADR are recorded.
- Once all of the disparate sound editorial elements are complete (and often premixed), the tracks are screened for client approval. Any go-backs should be completed before moving on to the next step.
- The tracks are transfered into a mix session, and the re-recording mixer begins mixing the tracks into the way they will finally sound. This is also where the different mix versions (eg. music and effects only, "clean" broadcast or in-flight movie versions) are prepared, as well as slightly different mixes tailored towards different release and distribution formats (eg. theatrical, home video, broadcast, surround, stereo).
- Once the mix is ready, the new tracks are synchronized with the final picture version (with all of the colour correction, effects, and titles), which is then used to create the marriage print or programme master tape.
This workflow will vary from project to project, but the basic elements remain fairly consistant. So, to actually answer your question: a dubbing mixer would probably ask for the original audio material in an OMF so that they would have more flexibility in remixing your tracks. If a mixed stereo file is the only option available, then they may go about it in any of a number of ways (including fader automation, normalization, and dynamics compression). The method really depends on the existing material, the desired result, and how far apart the two are. The principle that "what sounds good is good" is true if (and only if) the monitoring environment is exactly identical to the the audience's listenning environment. Since audience listenning environments vary so much, great care is taken in designing dubbing stages. An experienced mixer will also be aware the technical standards and variables associated with specific delivery methods, adjusting the mix with those in mind.
[alan smithee] "another way of asking the question is if i was to take my edit to dubbing mixer in a sound studio and ask him to increase the levels etc what method would he use"
If the overall quality was good and it just needed more volume, chances are he would raise a fader on the mixing console.
[alan smithee] "i'm trying to find out as much as i can about simple sound mixing so i can get decent results in the same way."
As has been explained sound mixing is not a simple task. There are many variables to take into consideration and guys who sit at a console on a dubbing stage have usually spent years perfecting there skills. It is not something you can teach someone in an online forum. That being said, I gave you the simplest solution to your situation in my first response.