Audio Trimmer & Converter
Cut MP3/WAV tracks visually, apply fade-in/out and normalization, and convert formats locally.
Voice Recorder Studio
Record voice memos with live spectrum, 80Hz rumble filter, and WAV/MP3 export.
Video to GIF & Trimmer
Trim MP4/WebM recordings and encode crisp animated GIFs in browser RAM.
Why Quick Audio Edits Shouldn't Require Desktop DAWs or Cloud Uploads
Whether you are trimming 5 seconds of dead air off the start of a podcast episode, normalizing a quiet voice memo before sending it to a client, or converting a raw WAV recording to a compact MP3—opening Audacity or uploading confidential interview recordings to a random online audio cutter is overkill and risky.
With the W3C Web Audio API (AudioContext + OfflineAudioContext), your browser already has a sample-accurate Digital Signal Processing (DSP) engine built in. The KitWharf Audio Trimmer & MP3/WAV Converter decodes, slices, fades, normalizes, and re-encodes audio 100% locally in RAM.
Web Audio DSP & MP3/WAV Encoding Pipeline
Web Audio API Decoding, OfflineAudioContext DSP & LAME MP3 Pipeline
From raw compressed MP3/M4A/WAV files to visual waveform scrubbing, linear fade envelopes, peak normalization, and binary export.
Preventing Audio Clicks & Mastering Voice Levels
- Zero-Crossing Fade Envelopes: Slicing an audio waveform mid-oscillation creates an abrupt DC jump that speakers reproduce as a sharp pop or click. Setting Fade In to
0.2sand Fade Out to0.3sramps the gain smoothly from silence. - Peak Normalization (
-0.2 dBFS): If a conference room recording is too quiet, enabling Peak Normalize scans every sample in your trimmed slice, calculates the true peak amplitude, and scales the entire track cleanly without clipping.
Deep Dive: How OfflineAudioContext & RIFF WAV / LAME MP3 Work in RAM
When you drop an MP3, M4A, or FLAC file into the KitWharf Audio Trimmer, your browser performs three distinct audio-engineering stages in memory:
1. Planar Float32 PCM Decoding
AudioContext.decodeAudioData() decompresses the container into raw uncompressed IEEE-754 32-bit floating-point PCM arrays (ranging from -1.0 to +1.0) per channel at the file's native sample rate (typically 44,100 Hz or 48,000 Hz).
2. Sample-Accurate Slicing & DSP Rendering
Rather than chopping compressed MP3 frames blindly (which corrupts frame headers and causes sync glitches), KitWharf slices the exact sample indices [startSample, endSample) and feeds them through an OfflineAudioContext graph:
- Peak Scan & Normalization: Every sample is scanned for
max(|sample|). If normalization is enabled, a uniform multiplier0.98 / peakboosts quiet recordings to-0.2 dBFSwithout clipping. - Linear Fade Envelopes:
gainNode.gain.linearRampToValueAtTime()smoothly ramps volume from0tototalGainover your Fade-In window, and back down to0over your Fade-Out window.
3. Binary Serialization (16-Bit RIFF WAV vs. LAME MP3)
- RIFF WAV Export: Writes a standard 44-byte RIFF/WAVE header followed by interleaved signed 16-bit little-endian integer PCM samples (
-32768to+32767). Zero generational quality loss. - LAME MP3 Export: Converts Float32 buffers to Int16 and feeds
1152-sample frames through@breezystack/lamejs(Mp3Encoder) at your chosen constant bitrate (128,192, or320 kbps).
Audio Format Selection Matrix
| Use Case | Recommended Format | Bitrate / Depth | Why |
|---|---|---|---|
| Video Editing (Premiere / DaVinci) | WAV | 16-bit PCM | Zero decode latency, frame-accurate timeline scrubbing |
| Podcasts & Voiceovers | MP3 | 192 kbps |
Universal player support, ~85% smaller than WAV |
| WhatsApp / Slack Voice Clips | MP3 | 128 kbps |
Fast mobile upload, crisp speech intelligibility |
| Music Loops & Mastering | WAV or MP3 320k | 16-bit / 320 kbps |
Preserves high-frequency shimmer and stereo imaging |