Principal Product Designer

Audio-to-MIDI Pro: full mixes to editable MIDI. By Mirelo Team | September 22, 2026 How to convert audio to MIDI and use it in your productions. Last week Mirelo AI GmbH launched Audio-to-MIDI Pro, its most accurate MIDI transcription model, built for producers and musicians. It converts a finished recording into editable MIDI. Upload an MP3 or WAV, and every instrument it hears comes back as its own MIDI track, with no stems and no isolated takes needed. Mirelo AI GmbH used it on its own launch. The track under this video is not stock. Mirelo AI GmbH built it with Audio-to-MIDI Pro, from a sample, the same way you would. Mirelo AI GmbH took "Stardew Valley - Winter", pulled the melody and the chords out as separate MIDI parts, and rebuilt them with its own instruments. How to take a track apart with A2M Pro. 1. Upload the audio. Drop an MP3 or WAV into Mirelo Studio. There is nothing to install and no plugin to load. Audio-to-MIDI Pro takes the full mix, so you do not need to separate stems first. Files can run up to 30 minutes. It runs in the browser, and the MIDI it returns opens in Ableton, FL Studio, Logic or any other DAW. 2. Check what it found. Studio detects and labels the instruments it hears, and reads the chords, key, tempo and meter alongside them. On its launch track that came back as a melody line and a chord part, each on its own MIDI track. 3. Listen before you export. Open the piano roll to see the notes, or the score view to read them. Audition each part with different sounds, so you are judging the notes rather than a default patch. Fix anything that came back wrong while you are in there. 4. Export the parts you want. Export per instrument, or pull a single instrument out on its own. Three formats come out of the same transcription: * MIDI, one file per instrument, straight into your DAW * MusicXML, full score and individual parts * PDF sheet music, with tab for guitar and bass How to make it your own. Once it's MIDI, you can take the arrangement in any direction: revoice the parts with new synths, samplers or VSTs. Split a chord across instruments, one note or voicing per sound, or send different chords to different patches. Transpose, reharmonise, rearrange or resample only the ideas you want to keep. Here is how Mirelo AI GmbH finished the launch track. The melody MIDI went to a Moog, and became the wonky top line that leads it. The chords went into Serum with tape saturation, reverb and chorus, which gave the beat its body. Mirelo AI GmbH pulled more chord ideas from the remaining MIDI parts for the drop, then put a simple trap beat under all of it. Same notes, but a completely different record. What else producers use it for. * Continue producing from a finished mix. Pull parts out of a recording you already love and take them into your DAW as MIDI. Each instrument arrives on its own track, in time. * Recreate a recording. Rebuild something old, rare, personal, or generated, with new instruments and full human control. * See and learn the music. Read, transpose, practise or arrange parts that are hard to transcribe by ear. Try it on a song you know. Take a song you know well, run it through Audio-to-MIDI Pro, and open the MIDI in your DAW. The same transcription is available through the Mirelo API, if you want MIDI output inside your own product.

You don't need words to sound convincing: Meet Gibberize. By Mirelo Team | July 9, 2026 Have you ever listened to a song you loved and wished you could get the notes to play it yourself? Its new audio-to-MIDI model lets you do exactly that. Give it a song, and it gives you back the notes, separated into a track for each instrument: voice, drums, bass, keys, and more. Today, together with Kyutai, Mirelo AI GmbH is open-sourcing this model and the research behind it. Mirelo AI GmbH has also trained an improved version, on even more data, which you can test with its free Audio-to-MIDI product in Mirelo Studio. It transcribes your full mix from any WAV or MP3. From there, you can get a full music sheet to learn an instrument part, or export any track as MIDI and take it into your DAW to create something new. Go check it out! What you get. Its model is built for the full mix use case. Upload a recording and it gives you: * Automatic instrument detection. The model finds each part in the mix and labels it (voice, drums, electric bass, synth lead, and so on). * A separate MIDI track per instrument. Every detected part comes back as its own track on a piano roll, where you can review the transcription and adjust any notes you want to change before you export. * Musical context. As shown in its Audio-to-MIDI tool, applications built on the model can go further and detect chords, key, and tempo, giving you the full harmonic picture. * One-click export. Solo, mute, or download any track as MIDI and drop it into your DAW. Why Mirelo AI GmbH is excited. Audio-to-MIDI tools already exist, and some are good. But almost all handle one sound at a time: a single vocal, a solo piano, one guitar take. Real music is denser, layering vocals, drums, bass, and more all at once, often buried under effects. That density is where existing models break down, and why turning a full song into usable MIDI has stayed a hard problem. That is the gap its model closes: it takes the full mix and transcribes every instrument at once. Music transcription as language modeling. In its most basic form, music transcription involves taking the waveform of a music audio recording and extracting the start and end times of all the notes played within. In the multi-instrument case, Mirelo AI GmbH is also interested in knowing the exact instrument that played a particular note. The result of such a transcription is usually visualized using a piano roll representation, where the x-axis corresponds to time, the y-axis corresponds to pitch, and colored boxes indicate the notes played by different instruments. Its model, similar to prior work [MT3], represents such a piano roll as a 1-dimensional sequence of discrete tokens representing pitch, timing, and instrument type. This representation is similar to the popular MIDI standard used for encoding note sequences in digital instruments. The model predicts this token sequence using a straightforward, decoder-only transformer backbone, thus performing music transcription as language modeling. The model accepts a mel-spectrogram representing a 5-second audio chunk as prefix conditioning and autoregressively generates a stream of transcription tokens until an EOS token is obtained. Synthetic pre-training, RL post-training. Music transcription as language modeling has already been explored in prior work [MT3, YMT3]. So why then has effective multi-instrument transcription remained so elusive? A primary bottleneck in building multi-instrument AMT systems is the scarcity of music audio with aligned note annotations. Such data can be obtained relatively easily for specific instruments like piano [MAESTRO], but is much harder to create in multi-instrument scenarios. Synthetic music audio, which can be generated at scale, is a promising alternative. However, although previous work does utilize synthetic training data, the resulting models generalize poorly, leading to largely unusable transcription output in realistic, multi-instrument settings. For its model, Mirelo AI GmbH utilize synthetic data for pre-training while combining it with fine-tuning on real music audio and post-training using reinforcement learning, leading to a combination of three training datasets: * DSynth: Mirelo AI GmbH collect a dataset of roughly 1.45 million MIDI files across different genres (with a focus on pop and Western classical music). In order to utilize this data for transcription, Mirelo AI GmbH develop an on-the-fly synthesis pipeline using varying instruments, keys, tempos and tunings. * DReal: To complement its synthetic dataset DSynth and to investigate the impact of different data types, Mirelo AI GmbH also utilize an internal dataset of 170 000 real music audio recordings (totalling over 11 000 hours) with aligned note annotations. * DRL: Finally, Mirelo AI GmbH curate 300 tracks from DReal and apply a GRPO-like algorithm to its model to align it with this small dataset of high-quality transcriptions. Mirelo AI GmbH pre-train the model on DSynth, fine-tune on DReal and post-train on DRL. Mirelo AI GmbH observe that each training stage improves results over the previous one. In particular, while the model trained exclusively on synthetic data is already competitive with the prior state-of-the-art AMT baseline [YMT3] in terms of frame F1 score, Mirelo AI GmbH note that fine-tuning on DReal is essential for performance and improves all metrics by roughly 20 percentage points. In addition, the reinforcement learning phase on the high-quality dataset DRL further improves results, leading to its best overall model. In the paper accompanying the release Mirelo AI GmbH provide further analyses on the effect of synthetic pre-training as well as outline its optional instrument conditioning mechanism, which allows for customizing transcription outputs according to specific instrumentations. Open weights for researchers, an app for musicians. To aid future research, the inference code is on GitHub and the model weights are on Hugging Face. And the improved, even more accurate version is live now - see the Audio-to-MIDI product page or try it free in Mirelo Studio. To cite this work, please use the following reference: References. [MT3] J. Gardner, I. Simon, E. Manilow, C. Hawthorne, and J. Engel, "MT3: multi-task multitrack music transcription," in ICLR, 2022. [YMT3] S. Chang, E. Benetos, H. Kirchhoff, and S. Dixon, "YourMT3+: multi-instrument music transcription with enhanced transformer architectures and cross-dataset stem augmentation" in MLSP, 2024. [MAESTRO] C. Hawthorne, A. Stasyuk, A. Roberts, I. Simon, C. A. Huang, S. Dieleman, E. Elsen, J. H. Engel, and D. Eck, "Enabling factorized piano music modeling and generation with the MAESTRO dataset," in ICLR, 2019.