Neural denoise Premium
RNNoise, a small neural network trained on speech, pulls hiss and room noise out of voice recordings. A strength slider blends the original back in, with the dry copy time-aligned.
Client-side audio · Browser, PWA, Android & iPhone · No upload farm
Live in browser Android on Google Play iPhone coming soon
Effects-first transcoder with a real signal chain — not another Play Store junk converter.
Drop a folder of audio, clean it with a neural denoiser, shape it with 10- and 20-band graphic EQs, compressor, distortion, delay and reverb, dial LUFS and radio simulation, pick a codec like you mean it (AAC-LC / HE-AAC, LAME MP3, MP2, FLAC, Opus, Ogg Vorbis, broadcast BWF / RF64, console ADPCM like PlayStation and Nintendo, telephony G.722 / G.726, and more), then batch-export a ZIP — entirely on your device, in the browser or the Android app. UI in 20 languages.
Free in the browser and on Android (the Android app is supported by ads). Premium ($1.99/month, in the iPhone and Android apps only) unlocks the specialty output formats and effects and removes the banner ad; they are not in the browser version. No upload — your audio stays on your device.
Most “audio converter” apps are format laundry: one slider, muddy defaults, and zero respect for loudness, psychoacoustics, or batch workflow.
What junk converters do
What chainDRiVE does
(cDRV batch).zipMedia queue, the effects rack (EQ, compressor, denoise, distortion, vocal cut and more), the Crystallizer, output format menu, import options, archive browser, ROM sample scanner, chain map, CHAiNAMP preview, job log, batch ZIP — swipe the rail.

















Every stage is intentional. Loudness first, creative FX in the middle, encode last.
Load common formats (MP3, M4A/AAC/MP4, FLAC, WAV, Opus/Ogg, WebM, AMR, AIFF…) and the specialty ones too (BWF, RF64, Wave64, CAF, VOC, 8SVX, MP1/MP2, Monkey's Audio, Musepack, WavPack, TTA, ALAC, Speex, DSD, WMA, RealAudio, DTS, ATRAC, tracker modules, MIDI/OPL, PlayStation, Nintendo, ADX, G.722/G.726, raw PCM). Resample and channel-map before any FX.
K-weighted integrated loudness with a target you set (broadcast-style −24…−6 LUFS range) and peak safety — not a dumb “normalize to 0 dBFS.”
The rack: vocal cut, neural denoise, silence removal, reverse, gain, 10- and 20-band graphic EQs, the Crystallizer, compressor, distortion, delay, reverb, Signal Strength makeup, the AM/FM radio simulator (hum, multipath, sferics, pilot…), Ear Protect shelf, fades and a look-ahead hard limiter. Every stage is off by default — drag the order, persist it. The Crystallizer is a spectral band replication (SBR) effect that rebuilds a natural top end above where narrow-band voice recordings cut off. See the effects rack.
Limiter closes the chain, then the encoder you actually chose. Multi-file jobs land in one (cDRV batch).zip.
Twelve stages that used to need a desktop editor, running on the device. Each one is switched off until you want it, and each one can be moved anywhere in the chain. In the browser you get LUFS, gain, the 10-band EQ, compressor, radio simulator, ear protection, fades and the hard limiter; vocal cut, denoise, silence removal, reverse, the 20-band EQ, the Crystallizer, distortion, delay and reverb are Premium, in the iPhone and Android apps only.
RNNoise, a small neural network trained on speech, pulls hiss and room noise out of voice recordings. A strength slider blends the original back in, with the dry copy time-aligned.
Centre-channel cancel for karaoke-style backing tracks, with a keep-bass crossover so the kick and bass guitar survive.
Ten octave bands or twenty third-octave bands, up to 24 dB each, with 30+ presets per EQ from Orchestral and Club Floor to Podcast Voice, Phone Line and Vintage Radio.
Full-range threshold, ratio, attack, release, knee and makeup controls, plus 14 presets from Lead Vocal Presence to Hard Ceiling.
Overdrive, tube, fuzz, hard clip, bit crusher, half-wave and full-wave rectifier, and band-pass distortion that drives only a chosen band.
Feedback delay and a generated-impulse reverb, both with their tails kept so the last echo is not cut off.
A look-ahead peak limiter you aim with ceiling, look-ahead and release. Gain eases down before the peak arrives.
Cut long gaps, play backwards, move the level by up to 24 dB either way, and fade the start and end.
Native-Android-style controls, ripples and navigation in the same amber on black, in 20 languages.
Encoded and decoded on your device. Every output can be previewed in the built-in player before you save it, and the specialty formats below can be loaded as inputs too. In the browser you can export WAV, AIFF, FLAC, MP3, AAC (incl. HE-AAC and M4A), Ogg Vorbis, Opus and WebM. The other output formats are Premium ($1.99/month), in the iPhone and Android apps only; opening files in any of these formats is free everywhere.
Perceptual codecs for music and everyday sharing.
Bit-exact when you do not want to throw anything away.
Studio, archive and legacy wrappers.
The broadcast workhorses, as .mp1 / .mp2 or inside WAV / BWF.
The audiophile and archive codecs, read and written on your device.
Export AC-3 and E-AC-3 surround-era codecs plus the MiniDisc-era ATRAC family.
Import-only: decoded on your device, then converted to anything above.
Sequenced music, rendered to audio so you can process and export it.
Export classic PC and console game audio formats.
Open an archive, pick a sound. Everything stays on your device. The archive browser lists the audio entries (filter, preview, select) and adds the ones you pick to the queue. The ROM sample scanner searches a cartridge image for sample data you can audition and add the same way.
4-bit ADPCM from consoles and arcades.
Low-bitrate speech codecs. G.726 is 8 kHz mono, G.722 is 16 kHz mono, OKI VOX is 8 kHz.
Formats that rely on a platform encoder (AAC via WebCodecs, AMR) appear only when your device can actually produce them; the software-implemented formats work everywhere.
If the UI only offers “High quality MP3,” it is not a transcoder — it is a toy.
AAC-LC (mono or stereo) plus stereo HE-AAC v1 (SBR) and HE-AAC v2 (SBR+PS) — as ADTS .aac or MPEG-4 .m4a. LC uses native WebCodecs with an fdk-aac (WASM) fallback; HE-AAC runs on fdk-aac (WASM).
CBR/VBR, HQ toggle, and six psychoacoustic/stereo profiles (LAME Default, FHG Narrow, Music, Speech, Wide Stereo, MS Joint Only), with adjustable stereo width and Force-MS joint stereo — knobs for people who hear the difference.
FLAC, AIFF, AU/SND (Sun Audio), and WAV PCM in 16- or 24-bit integer and 32-bit float — bit-exact when you do not want to throw anything away.
Opus (.opus) with Audio / VoIP / Low-Delay application modes and complexity 0–10, plus Ogg Vorbis and WebM (Opus) for the open-web stack.
AMR-NB (8 kHz) and AMR-WB (16 kHz), WAV IMA & Microsoft ADPCM, G.711 A-law / μ-law, and GSM 6.10 — the low-bitrate and legacy codecs most converters skip.
Sample rates from 8 kHz telephony up through 48 kHz; mono, stereo, or joint stereo where the codec allows; size estimates before you commit.
Formats appear when the device can actually encode them — no fake “Supported” badges for codecs the browser cannot ship.
CSP locks the app to itself. Processing uses local OfflineAudioContext and workers — your stems are not a SaaS upload.
Terrestrial AM/FM character printed to a file — the Dialer's Lab radio-chain work, rendered offline. Pick Off, AM, or FM, dial Signal Strength, and toggle each artifact on or off.
Eight switchable artifacts: mains hum, power-line buzz, background hiss, heterodyne whistle, crackle & pops, static crashes (sferics), selective fading, and carrier flutter.
Eight switchable artifacts: background hiss, multipath dips, crackle & rasp, power-tool RFI, stereo separation, ceiling whistle, stereo pilot tone, and program flutter.
Queue a folder, set one chain + one encode profile, run once — get a ZIP, not twelve mystery downloads.
Drag-drop, select, process all. Same loudness and FX recipe across the set.
Terminal-style process log so you can see decode → LUFS → FX → encode, and abort mid-run.
(cDRV batch).zip packaging for the whole run — built for people who convert libraries, not single clips.
Browser, installable PWA and Android are live; the iPhone app is coming soon. The browser version is free and has the everyday formats and effects. Premium ($1.99/month) adds the specialty output formats and effects and removes the ad banner, and is available in the iPhone and Android apps only. Same studio as Global Chaos CB Radio and KAOS-FM.
Full workbench on desktop, phone-first chrome on mobile, installable as a PWA. No account, nothing uploaded. Premium formats and effects are not in the browser version.
The same engine in an app, supported by ads, with the option to subscribe to Premium (all formats and effects, no banner). Big batches keep running with a foreground-service progress notification, so converting a folder does not stall when you switch apps.
The interface ships in 20 languages: Arabic, Chinese, Dutch, English, Filipino, French, German, Hindi, Indonesian, Italian, Japanese, Korean, Polish, Portuguese, Russian, Spanish, Thai, Turkish, Ukrainian and Vietnamese.
Free, on-device, no signup — open it in your browser or get the Android app, and drop in a file. Want every format and effect? Premium is in the iPhone and Android apps.