Differences

This shows you the differences between two versions of the page.

--- software:audition [2023/07/13 15:52] – [Loudness] mh
+++ software:audition [2025/08/21 16:43] (current) – [Personal Workflow] mh
@@ Line 11: / Line 11: @@
   * In the **Preferences > Audio Hardware** Set the master clock to reflect whatever the main use of the software is at the moment, so when doing a lot of recording set it to the default input but when mastering set it to the major output.
+----
 ===== Loudness =====
 When Mastering, it's important to know what loudness your aiming for. This will depend on the content you are producing. Songs will not be mastered the same way as podcasts, or video.
-LUFS (Loudness Units relative to Full Scale) are used as a standardized way of expressing loudness. It’s a measurement of how loud the peaks are compared to the average level. So the more dynamic the track, the higher the number. The less dynamic the track, the lower the number.
+LUFS (Loudness Units Full Scale) are used as a standardized way of expressing loudness. It’s a measurement of how loud the peaks are compared to the average level. So the more dynamic the track, the higher the number. The less dynamic the track, the lower the number.
   * **Integrated LUFS** are a measurement taken usually over the entire sample.
@@ Line 25: / Line 26: @@
 The **[[https://www.loudnesspenalty.com|Loudness Penalty Analyzer]]** can be used to help guide the mastering process for loudness normalization.
-Podcasting generally targets -14 LUFS.
+Podcasting generally targets -14 LUFS. Every online streaming platform operates with its own Distribution Loudness reference level, generally around -14 LUFS but there is no standard. If your audio is above that level, they will all tune it down. If it is under, some platforms will tune it up (Spotify for example does) and some won't (YouTube for example doesn't).
 === Additional Ressources on LUFS ===
   * [[https://www.mixinglessons.com/lufs-loudness-unit-full-scale/|Mixing Lessons Article on LUFS]]
+----
+===== Voice Mastering Workflow =====
+==== Personal Workflow ====
+Inspired by Mark Russel's workflow described below, I apply on my voice recording tracks the following stack :
+=== 1.Parametric EQ ===
+{{:software:parametric-eq.png?direct&400|}}
+=== 2.DeEsser ===
+{{:software:de-esser.png?direct&400|}}
+=== 3.Dynamics ===
+{{:software:dynamics.png?direct&400|}}
+=== 4.Mastering ===
+//Applied on the Mix channel and output individual tracks to Mix//
+{{:software:mastering.png?direct&400|}}
+==== Mark Russel Workflow ====
+Start with **Equalizer** and use a low pass filter to cut off anything below 50 or 80 or 100 Hz, depends on the voice and personal preference.
+Wobble around in the EQ if needed to dim off inelegant frequencies.
+Add a **High Pass gain** to boost the highs a little.
+Then add **Dynamics** to add both a **Compressor** and a **Limiter** in one bundle Fx.
+Set the compressor and choose a threshold that works well for the voice recording. Use 2 or 4, Mike goes for 2, I settled for 2.5.
+Then add the limiter and set a threshold to a dB value where you would like to limit the audio (-3dB or -1.5dB works well)
+Then adjust the make-up value so that it tickles the limiter (red square) on voice playback.
+Finally, add a Mastering Fx to the general Mix channel and use the preset "Sublte Clarity" which works pretty well.
+Tweak it a little if necessary.
+----
+==== Alternative Workflow ====
+Start by removing ambient noise, by using a sample silence in the recording.
+Use a **Compressor**. Look for the average peaks in the signal, (not the high ones, but the range that is consistently being hit on average). Set that as the threshold level of the compressor.
+A good ratio for dialog is 4:1.
+Attack and Release can be left to 10/100 ms.
+Output gain can be used in the compressor to generally get the levels back up after they have been compressed but this can also be done afterwards with another effect if left to 0dB.
+Use a **DeEsser**. Men are generally between 5-8 kHz and Women generally between 6-7 and 10 KHz.
+Isolate a loop in the dialog with a lot of sibilance sounds.
+Open the de-esser and find a center frequency and a bandwidth that work pretty well. Examin it with the output sibilance only. Then tweak to find a good threshold to remove the sibilance sounds without denaturing the voice too much.
+Use a **Parametric EQ**. Treat the low-mids first before boosting the highs.
+Apply a low pass filter and make the curve a little severe so that it doesn't remove too much the low frequencies but cleans up the deepness.
+Then lower the low-mids with a high cue generally where the curve is above 0.
+Then apply a little boos to the highs.
+Use a **Hard Limiter** to boost the overall signal.
+Look at the peaks and boost them to -3dB approx. by using the input boost. If the peaks are at -9 add a 6dB boost for example.
+Maximum amp can be set to -1dB for safety.
+----
 ===== Keyboard Shortcuts =====
-==== Zoom ====
+=== Zoom ===
   * ''+/- (KEYPAD)'' : Zoom In/Out by one step
-==== Multitrack/Waveform ====
+=== Multitrack/Waveform ===
   * ''KEY_PAD 0'' : Toggle back from Waveform/Spectral view to Multitrack view
   * ''KEY_PAD 9'' : Toggle from Multitrack view to Waveform/Spectral view of selected clip
-== Playback ====
+=== Playback ===
   * ''HOME'' : Set playhead at the beggining of the session
@@ Line 51: / Line 133: @@
   * ''L'' : Read forward from playhead (multiple hits increase speed of playback)
+----
+===== Ressources =====
+  * **[[https://www.youtube.com/channel/UCYEiWHoP2IHHgjOapTAYMQw|Mike Russel's YouTube Channel]]**
+  * **[[https://podcast.adobe.com/en/miccheck|Adobe Podcast Mic Check]]**