Free
$0freeFull instrument and your own samples, with the Basic sound engine and 20 one-time generation credits (20 a month for the first 200 users). Official source 1 ↗
Website-checked tool profile
2G
A tiny AI sampler that turns text descriptions into playable sounds with an early-2000s phone-speaker character.
Sources checked
- Pricing
- Freemium. The free version has the full instrument and your own samples, plus 20 one-time generation credits (20 a month for the first 200 users). Studio costs $10 a month for 200 generations a month with a newer Studio sound engine, and has a 7-day free trial. One credit is one successful generation. Official source 1 ↗
- Available as
- Desktop instrument for Windows 10 or later (VST3 and standalone) and macOS 11 or later on Apple Silicon or Intel (AU, VST3 and standalone). Official source 1 ↗
- Primary group
- Music & Audio
Also consider
See all 6 similar tools ↓- SFX EngineAI sound effect generator that turns text prompts into royalty-free SFX, with background music, DJ transitions and a stem splitter.Paid
- Tera StudioClone your singing voice from about a minute of audio and hear any song sung back in your own voice.Free plan
- Fish AudioVoice AI platform for expressive text-to-speech, voice cloning and transcription, with open-weight models and a pay-as-you-go API.Free plan
What it does
2G is a small sampler instrument from Scrap Brain Audio for Windows and macOS, styled after early-2000s mobile phones. You type a description of a sound, its Neural Synthesis feature generates it with AI, and 2G then finds loop points and the root note so you can play the result on a keyboard. It runs standalone or as a VST3 or AU plug-in, and it can also load your own samples without using credits.
Official source 1 ↗Features
- Text-to-sound Neural Synthesis. Describe the sound you want in words and 2G generates it with AI, then loads it as a playable sample. An optional prompt-enhance setting can rewrite your description before generation. Official source 1 ↗
- Automatic loop finding. 2G compares timbre, volume and vibrato to find loop points with few abrupt shifts, crossfades them and offers up to three candidates, so a held key sustains the sound. Official source 1 ↗
- Root note detection. It measures pitch across the sample, corrects fine-tuning on playback, and marks samples with no clear pitch as unpitched. Official source 1 ↗
- Sample shaping controls. Trim the sample and shape it with Attack, Release, Brightness, Detune, Noise, Quality and Formant controls. Official source 1 ↗
- Phone-speaker simulation. An optional model of a small mobile speaker adds its narrow frequency range, resonances and 16 kHz mono character, with voice-coil heating that compresses sustained loud playing. One switch bypasses it. Official source 1 ↗
- Use your own samples offline. Import WAV, AIFF, FLAC or MP3 files up to 20 MB and 10 seconds. Editing and playing local samples uses no credits and works offline; only generation needs an internet connection. Official source 1 ↗
Pricing & free limits
Studio
$10per month, 7-day free trialFull instrument with the Studio sound engine and 200 generations a month. Official source 1 ↗
Billing details
One credit equals one successful generation. Studio credits do not roll over, and after cancelling, generation continues until the end of the paid period; the local sampler and saved sounds stay available. Official source 1 ↗
Free access & limits
20 generation credits in total (20 a month only for the first 200 users) on the Basic sound engine; importing and playing your own samples is unlimited. Official source 1 ↗
Best for & limitations
Generating a lo-fi, phone-speaker-style synth or texture from a text prompt and playing it as an instrument in a DAW via VST3 or AU. Official source 1 ↗
Turning your own short one-shot recordings into looped, tuned playable instruments without spending credits. Official source 1 ↗
Making royalty-free sounds for commercial tracks. Official source 1 ↗
- Very small free allowance. After the first 200 users, the free version gets only 20 generation credits in total; regular generation needs Studio at $10 a month. Official source 1 ↗
- Short samples, lo-fi by design. Imported samples are limited to 10 seconds and 20 MB, and the instrument's character is 16 kHz mono unless you bypass the phone speaker. Official source 1 ↗
- Prompts are sent to a provider and kept. Generation needs internet. Your prompts go to an unnamed sound-generation provider and are kept for up to 90 days; the site names no company behind Scrap Brain Audio. Official source 1 ↗
- No resale as sample packs. Generated sounds can be used in music, but not repackaged or sold as standalone samples or libraries. Official source 1 ↗
Similar tools
Tools like 2G
Published tools that share its categories.
SFX EngineAI sound effect generator that turns text prompts into royalty-free SFX, with background music, DJ transitions and a stem splitter.
Visit Tera StudioClone your singing voice from about a minute of audio and hear any song sung back in your own voice.
Visit Fish AudioVoice AI platform for expressive text-to-speech, voice cloning and transcription, with open-weight models and a pay-as-you-go API.
Visit 
YuE2 StudioFree Windows desktop studio that runs the open YuE2 song model on your GPU: full songs with vocals and an editable score, covers, MIDI and LoRA training.
Visit 
Vidu Q4Vidu’s preview video model that turns images or references into up to 16-second, 4K clips with native audio.
Visit 
SpliceRoyalty-free sample library with 3M+ sounds, presets and MIDI, AI Search with Sound, in-DAW integrations and rent-to-own plugins.
Visit