
Keep the selected channels identical when comparing Audacity Plot Spectrum measurements. The analysis sums selected channels, so selecting a stereo pair instead of a single channel changes the analysis input. A different plotted level is not automatically evidence that an equalizer or export changed the source.
Key Takeaways
- Record which channels and passage the plot analyzes.
- Compare like-for-like selections with the same settings.
- Use listening and appropriate output checks alongside the frequency plot.
Define the comparison before opening the analyzer
Write the question in measurable terms: whether a low-frequency tone is present, whether processing changed a frequency region, or whether two microphones differ. A plot of different content cannot reliably answer a before-and-after processing question.
Audacity's Plot Spectrum reference describes averaging analysis blocks and summing selected channels. It also documents Replot after a selection change. Keep the selection and settings visible in your notes rather than comparing screenshots whose inputs are unknown.
Make the analysis conditions match
- Preserve the original and processed examples.
- Select the same-duration, corresponding passage in each.
- Choose the same channel scope, such as the intended microphone alone.
- Use the same algorithm, window size, and axis settings.
- Replot after changing the selection, and label the results with their actual inputs.
If the source is a stereo recording with different material on each side, decide whether the task concerns one microphone or the combined delivery. That choice should follow the question, not whichever plot looks more dramatic.
Work through a two-microphone example
Imagine that one channel contains a close microphone and the other contains room sound. A plot of only the close microphone is being compared with a plot of both channels after an edit. The extra room contribution could change the result even without the processing being investigated.
Repeat the comparison using matching inputs. Then listen to the relevant speech and pauses. Acceptance means that the observation supports the stated question under known conditions, rather than merely showing that two images look different. This is an illustrative analysis procedure, not original measurement data.
Know what the plot cannot certify
A spectrum averages information over the selected passage. It does not by itself certify timeline synchronization, speech intelligibility, or a correct deliverable channel layout. Use separate checks for those properties.
For identifying recorded microphone channels, the two-microphone source guide explains the content distinction through Premiere. For treatment after identifying a sustained tone, see the narrow-notch hum guide. A visually changed plot alone is insufficient evidence of a successful repair.
Related tasks
- How Do I Edit Two Microphones Recorded on One Stereo Clip Separately in Premiere?
- How Do I Remove Electrical Hum Without Thinning the Whole Voice?
Sources and scope
- Audacity Manual: Plot Spectrum: Spectrum analyzes averaged blocks; sums selected channels; channel selection affects plotted level; Replot updates changed selections.
Prepared by Imagild Editorial, the company editorial team responsible for these guides. AI assisted research and drafting. This is documentation-based guidance, not a claim of hands-on testing or independent review. Report a correction.
Cover: Pignatta / Pixabay, Content License. The city photograph is an illustrative source asset; it does not show this application or a test result.