StemSplitterLab
Split a song into 6 stems in your browser and practice along
This is 1 of 124 launches in music production and audio software — see how it stacks up on momentum and crowding →
140 other launches read as similar to this one →
Details
- External ID
- 1266172
- Source
- PH
- Company
- —
- Product
- StemSplitterLab
- Website domain
- producthunt.com
- Launched
- Oct. 1, 2026
- Cohort
- —
- Upvotes
- 1
- Upvotes percentile
- 0.3309652131308182
- Tags
- Music, Education, Audio
- Fetched at
- Oct. 2, 2026, 1:01 a.m.
- Updated at
- Oct. 2, 2026, 1:01 a.m.
Description
StemSplitterLab splits a song into six stems (vocals, drums, bass, guitar, piano, other) right in your browser. Your audio is not uploaded. Mute the part you play, then loop a phrase, slow it down without changing pitch, transpose, and practice with a metronome and count-in. Export the mix as WAV or MP3, or a single stem as WAV. Free, no account. A 136 MB model downloads once after the page opens and is cached. Guitar and piano stems are less clean.
Enrichment
- Theme
- music production and audio software
- Vertical
- Media & entertainment
- Function
- Content generation
- Audience
- Prosumer
- AI stance
- AI-native
- Project type
- Commercial product
- Normalized one-liner
- in-browser audio stem separation tool for musicians
- Manually corrected
- False
Could you build this?
Partial Running a high-quality 6-stem audio separation model locally in the browser requires WebAssembly/WebGPU, ONNX Runtime Web, and optimized client-side audio DSP pipelines that are technically challenging to execute smoothly.
What it would actually take: Requires porting complex deep learning music demixing models (like Demucs v4 or HTDemucs) to ONNX/WebGPU with aggressive quantization to fit into browser memory limits. Pair with Web Audio API for playback manipulation (time-stretching via SoundTouch.js, pitch-preserving playback, Web Audio gain nodes). The hardest part is client-side memory management and WebGPU compute shaders to prevent browser tab crashes during heavy multi-channel audio inference.
Competitors
Other products that read as similar to this one — 140 launches clear the similarity bar, closest 8 shown.
Attention rank: #52 of 141 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 330 days after the earliest competitor.
- StemConsole · ph · 2026-09-23 · 1 upvotes · similarity 0.82
- AudioPrism · ph · 2026-09-17 · 2 upvotes · similarity 0.61
- OffgridStem · ph · 2026-09-16 · 1 upvotes · similarity 0.59
- Song Practice AI · ph · 2026-09-15 · 1 upvotes · similarity 0.57
- Demucs music stem separator rewritten in Rust · hn · 2026-03-03 · 19 upvotes · similarity 0.56
- StemSplitter Studio · ph · 2026-09-16 · 2 upvotes · similarity 0.56
- Noize · ph · 2026-09-18 · 1 upvotes · similarity 0.53
- Practice Room · ph · 2026-09-09 · 2 upvotes · similarity 0.51
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a content generation tool for Government yet.