Your file is decoded to raw samples in the browser. This channel converter handles three workflows: Mono → stereo copies the single channel to both left and right (dual mono); the optional width control adds a short Haas delay to one channel so it feels wider — a psychoacoustic trick, not real stereo. Stereo → mono averages the channels, (L+R)/2. Adjust width uses mid/side processing: it splits the signal into mid M=(L+R)/2 and side S=(L−R)/2, scales the side by your percentage (0% collapses to mono, 100% is unchanged, 200% is twice as wide), then rebuilds L=M+S, R=M−S. The result is exported as a lossless 16-bit WAV. Nothing is uploaded.
Real stereo carries information captured by two microphones (or panned in a DAW mix). No tool can invent that from a mono recording — widening only simulates space using psychoacoustic cues. If you’ll re-sum the audio to mono later (some phones, club PAs, mono Bluetooth speakers), heavy Haas widening can partially cancel, so keep it modest. The mono-compatibility issue is especially relevant for broadcast, podcast production, and social-media video, where many listeners use a single speaker. For a lossless WAV that you then want to analyse spectrally, the audio spectrum analyzer can show you whether the widening has introduced comb-filtering artifacts in the high-frequency range. For MP3 output, re-encode the WAV in a desktop app such as Audacity, Adobe Audition, or FFmpeg.
Is my file uploaded?
No. Decoding, preview and conversion all happen in your browser; the audio never leaves your device.
Can this make real stereo from a mono recording?
No. True stereo needs two channels captured (or mixed) separately. This duplicates the mono signal and can simulate width with a Haas delay, but it can’t recover spatial detail that was never there.
How does “Adjust stereo width” work?
Mid/side: it scales the difference between the channels. 0% = mono, 100% = original, up to 200% = wider. Above 100% can clip — peaks are clamped on export.
Why is the output WAV, not MP3?
Browsers can’t encode MP3/AAC from JavaScript without a licensed encoder. The result is exported as lossless WAV; re-encode it to MP3 in a desktop app if needed.
My mono file already plays in both speakers — why convert it?
Some players up-mix mono on playback, but the file itself is still single-channel. Converting writes an actual 2-channel file, which some apps, uploads, or editors require.
What is a Haas delay and how much widening should I apply?
The Haas effect (also called the precedence effect) is a psychoacoustic phenomenon: when the same sound arrives at each ear with a small delay (roughly 1–40 ms), the brain perceives it as coming from the earlier-arriving side, creating a sense of spatial width. This tool adds a short delay to one channel to simulate stereo space from a mono source. For most use cases, a moderate width setting (around 30–50%) gives a convincingly wider sound without audible coloration. At high widening values, you may hear a slight hollow or "phasey" quality, especially on headphones. Keep widening below 50% if the file will be used anywhere that might sum to mono.
Will the converted file be mono-compatible? Can I use it on YouTube, TikTok, or podcasts?
A simple dual-mono conversion (0% width) is fully mono-compatible — left and right channels are identical, so summing them produces the original signal without any cancellation. Widening above 0% introduces inter-channel differences: when the two channels are summed to mono (as they are on some phones and streaming devices), the Haas-delay component partially cancels, which can cause a subtle level dip or tonal change. For YouTube and podcast uploads, dual mono or light widening (under 30%) is safest. For TikTok, Instagram, and similar platforms that often play back on mono Bluetooth speakers, stick to dual mono or test the output after summing. You can use the
audio trim tool to isolate a short test section before committing to a full render.
What is mid/side (M/S) processing and when should I use it?
Mid/side processing splits a stereo signal into two components: the mid (M = L+R, the part both channels share — essentially the mono content) and the side (S = L−R, the part that differs between channels — the stereo image). By scaling the side component independently, you can widen or narrow the stereo image without affecting the center. At 0% side the output collapses to mono; at 100% the image is unchanged; above 100% the image is artificially widened. This tool's "Adjust stereo width" mode uses M/S to non-destructively reshape an existing stereo file. It is useful for narrowing an overly wide mix for broadcast, or for gently expanding a close-mic recording where both channels captured nearly the same signal. Note that widening an already-narrow source above 100% amplifies any small channel differences, which may reveal background noise or artifacts.