Pod Level
Automatically balance loud and quiet speakers in your podcast
Fix recordings where one speaker is louder than another. Pod Level adjusts the changing levels of a conversation, turn by turn, instead of simply normalizing the whole file.
- Free
- No signup
- Runs in your browser
- Audio never leaves your device
Choose an episode
This tool requires a supported browser with local WebCodecs audio, worker, WebAssembly, and output-saving support.
Your audio stays on this device. Nothing is uploaded.
Works in current iPhone Safari, Chrome on iPhone, and desktop Chrome or Edge. The original file is opened read-only; leveling writes a new file.
Review the local job
- Duration
- —
- Channels
- —
- Sample rate
- —
- Input size
- —
- Proposed format
- —
- Estimated output
- —
This browser cannot check free space at the destination. Confirm that the selected disk has room for the estimated output.
Processing makes a new file. It never changes the source.
Leveling locally
Your audio stays on this device throughout analysis and rendering.
Preparing local processing…
Timestamp 0:00 / 0:00
Speed Waiting for a stable reading
ETA Calculating
Speech turns 0
Leveled file saved
- Input loudness
- —
- Output loudness (before encoding)
- —
- Input peak
- —
- Output true peak (before encoding)
- —
- Applied turn gain
- —
- Turns adjusted
- —
- Processing time
- —
- Average speed
- —
- Output format
- —
- Output size
- —
Before and after
Preview a few short handoffs with both versions starting from the same point.
Processing canceled
No audio was committed. On platforms that create a destination before the transactional write begins, an empty destination file can remain and may be deleted safely.
The file could not be leveled
Hear the difference
A scripted exchange between two synthetic voices, with the guest 12 dB quieter than the host. The second clip is the first one processed by Pod Level; nothing else was changed.
Why ordinary normalization isn’t enough
Normalizing applies one gain to the whole recording. If the host is loud and the guest is quiet, both move together and the gap between them stays. Pod Level measures each speech turn, brings quiet and loud turns toward the recording’s typical speech level, then normalizes the result and limits true peaks. Within a turn the gain is steady, so how people speak is kept. How speaker balancing works