Get AI feedback on your
AI production mentor for electronic music producers. Upload a bounce from any DAW and get feedback that knows your mix inside out. Working in Ableton? Add your .als and the feedback names your actual tracks and devices.
Free analysis. Unlock the full report for €5.95 or go Pro.
We read your audio AND your Ableton project. That's how the feedback gets specific.
How it works
From upload to actionable feedback in under two minutes.
Drop your track
Drop a bounce from any DAW. Working in Ableton? Add your .als for the deepest feedback.
Deep analysis against your genre
Mix balance, dynamics and groove, scored against your genre profile. Arrangement gets written feedback, not a score.
Key insights at a glance
The most important findings from your analysis, sorted by severity, each with the measurement behind it.
'Sub Bass' is still ringing when the next 'Kick' hits, so the low end smears. Shorten the sub notes or duck them under the kick.
'Pad' builds up in the low-mids and clouds the bass. Thin it out there before you touch the master.
'Clap' and 'Lead' hit the same range on the backbeat. A small cut in 'Lead' gives the clap more room.
Talk to your Sensei
Ask anything about your track. The Sensei answers from your own analysis data.
For when you're stuck and don't know why.
You can probably relate. Six hours into a session, your ears are tired, and something about the track isn't sitting right, but you can't pin down what. The buildup feels off, or the drop lands flat, or the mix is muddy in a way you swear wasn't there yesterday. You alternate between soloing channels and asking yourself whether you're being too critical, or whether it really is off. Your reference track gave you a target, not the steps to get there.
That's what TrackSensei was built for. We open your .als project file and your audio bounce together, and analyse both files: the structure of your project, the plugin chain on each channel, your routing and sidechains, the MIDI behind every part. So when we say "the pad is fighting the lead at 2:34", we mean exactly that pad on exactly that bar. Notes you can act on, in producer language, that point at the specific elements in your specific track.
And it adapts to what you actually make. Hypnotic techno isn't peak time. Peak time isn't hardstyle. The feedback knows the difference. The more context you give it (your reference track, your plugin chain, your level as a producer), the sharper it gets. Your taste, your tools, your direction. We just help you hear it.
The analysis is free. You get the scores, how your track sits against your reference, and every issue we found, each with its severity and the measurement behind it. The part you unlock per track is the work on top of that: why each issue matters, how to fix it, what to expect afterwards, the deep dives, and a chat that stays on your track.
Everything your track needs
Ableton Project Parsing
We read your .als file: tracks, devices, MIDI, automation, sends, everything.
How .als analysis worksAudio Analysis
LUFS, PLR, PSR, Bark-weighted spectrum, multiband stereo correlation, the metrics real mastering engineers use.
Production Scores
Mix balance, dynamics and groove, each scored 0-100 against your genre profile (dynamics from the mastering phase on). Arrangement gets written feedback instead of a number.
AI Chat Mentor
Ask follow-up questions. Your Sensei knows every detail of your track.
MIDI & Groove Analysis
Chord detection, swing analysis, velocity curves, and pattern repetition insights.
Genre-Specific Feedback
12 main genres, 71 subgenre profiles, one measurement engine. Industrial Techno scores against different targets than Minimal Techno, Gabber against different targets than Frenchcore, Phonk against different targets than Trap. Techno and house targets come straight from 563 released tracks; the other genres are built from in-depth genre research and get sharper as more tracks come in. Pick your subgenre and the scoring adapts.
AI mix feedback for your genre
Genre-aware feedback tuned to how each style is actually mixed.
Stem Interaction Analysis
Upload 2–4 stems. Get pinpoint diagnostic feedback on kick-bass clash, sidechain coherence, fundamental frequency tuning, and frequency masking between specific elements.
Data helps. Your ears decide.
Music lives in feeling, in groove, in the moments that give you goosebumps. No algorithm replaces that. We point you where to listen. The final call is yours.
Pricing
Three ways in: a free analysis every month, one Release Check for a single track, or Pro for a monthly budget.
Free
Try it out, no strings attached
- 2 free analyses per month
- Scores and how you compare to the reference
- Every issue we found, with severity and measured value
- The full fix for your top issue
- Compare your latest two versions
- 3 chat questions per month
Release Check
Unlock the full report for one track
- Every fix explained: why it matters, how to fix it, expected effect
- Pillar deep dives and issue threads
- 20 mentor chat messages on our best model
- One extra version to check your fix
- Report readable for a year
Pro
For producers who want to level up
- 150,000 feedback tokens / month
- ≈ 15 full track journeys (or 50+ quick analyses)
- Full report: every fix explained, deep dives, version compare
Ultimate
No limits, maximum growth
- 400,000 feedback tokens / month
- ≈ 40 full track journeys (or 130+ quick analyses)
- Longer, more detailed mentor answers
- Stems analysis included
- Boosters available for extra tokens
A track journey is roughly 1 analysis plus 6 chats (≈ 10,000 tokens). A quick analysis without follow-up chats runs about 3,000 tokens. Tokens are flexible, so you can chat more and analyze less, or the other way around. Prices include VAT where it applies.
A Release Check is one payment for one track. Chat, deep dives and the extra version are open for 14 days; the report stays readable for a year.
Common questions
Audio: WAV, MP3, FLAC, AIFF, OGG, bounced from any DAW. Every analysis starts from your audio. Project files: Ableton Live .als. Add one to your bounce for the deepest feedback, which names your actual tracks and devices.
ChatGPT has never heard your track. TrackSensei analyzes your audio against genre-specific targets, then the AI interprets those numbers for your specific mix. Concrete and actionable, not general.
No. We use Anthropic's Claude API which does not train on API data. Your music, projects, and conversations are never used to improve any AI model.
Yes. Every account gets 2 free analyses a month: scores, how you compare to the reference, and every issue we found with its severity and measured value. You pay only when you want the fixes explained.
12 main genres in the upload dropdown, each with dedicated subgenre profiles for sharper feedback. Pick the closest match to your track and the scoring adapts to your selection:
- Techno: Peak-time Techno, Hard Techno, Industrial Techno, Schranz, Raw Techno, Melodic Techno, Dub Techno, Detroit Techno, Minimal Techno, Acid Techno
- House: Deep House, Tech House, Disco House, Bass House, Amapiano, UK Garage, Progressive House, Melodic House, Organic House, Afro House, Future House, Acid House
- Trance: Progressive Trance, Uplifting Trance, Psytrance, Goa Trance, Full-On, Vocal Trance, Hard Trance
- Drum & Bass: Liquid DnB, Neurofunk, Jungle, Halftime, Jump Up
- Dubstep: Riddim, Melodic Dubstep, Future Bass, Tearout, Brostep, Color Bass
- Hip-Hop: Boom Bap, Trap, UK Drill, Chicago Drill, Lo-Fi Hip Hop, Phonk, Cloud Rap, Emo Rap
- Hardstyle: Rawstyle, Euphoric Hardstyle, Xtra Raw
- Hardcore: Gabber, Uptempo Hardcore, Frenchcore, Happy Hardcore, UK Hardcore, Speedcore, Terror
- Rock: Punk, Post-Rock, Metal, Metalcore
- Ambient: Downtempo, Drone, Dark Ambient, Cinematic
- Pop: Synthpop, Dance Pop, K-Pop, J-Pop
- Breaks: Big Beat
Each subgenre has its own LUFS targets, spectral balance, stereo rules, BPM range, and arrangement expectations. The scoring engine treats Industrial Techno, Dub Techno and Minimal Techno as three distinct things, not as one generic Techno.
A one-off payment of €5.95 for one track. It opens every fix explanation, the pillar deep dives, the issue threads and mentor chat on our best model, plus one extra version to check your fix. Chat, deep dives and that extra version stay open for 14 days, and the report stays readable for a year.
Ready to level up your tracks.
Start with a free analysis: scores, how you compare to the reference, and every issue we found. Unlock the full report for €5.95 when you want the fixes.
Get Started Free