Somali Music AI · The Golden Age of Qaraami
When my teacher and I went looking for Somali sheet music, there was none. So I built Somali Music AI — combining artificial intelligence and traditional music training so people and machines can finally generate, transcribe, and understand Somali music.
From a recording to a page
Press play. The engraving follows the audio note by note — the same binding the studio uses on full recordings.
Qaraami phrase in A pentatonic
Research demos
Short audio results from the project, each with what it is and how it was made. More are added as the research moves.
The melody was engraved as sheet music, parsed from its MIDI into 126 timed notes, and is performed here live in your browser by the platform’s own instruments: a plucked-string oud voice (a doubled course, two strings a few cents apart) over a durbaan frame-drum groove. Nothing is a recording — every note is synthesised the moment you press play.
Composed by Abdillahi Qarshe, 1950s · Sheet music by Khalid Ibrahim
A screen recording of the QaraamiGenAI studio — composing with the oud and durbaan instruments — will sit here.
Maqalka Qaraamiga · Preserved Master Sessions
Seven authentic archival takes from the private Qaraami collection. Unhurried solo kaban (oud) improvisations, classical heelo melodies, and poetic balwo verses in the golden Radio Hargeisa tradition, preserved with their natural acoustic warmth.
Kaydka Hiddaha Soomaaliyeed
“Fanka Iyo Suugaantu waa ilaaliyaha Afkeena Hooyo.”
Shanqadha xadhkaha kabanka iyo codkii qaraamiga — waa dhaxal ma-guuraan ah oo facba fac u gudbinaya kaydka suugaanta iyo dhaqanka Soomaaliyeed ee Raadiyaha Hargeysa ku bilaabmay.
A short heelo in the authentic qaraami manner, oud-led.
Qaraami song with the oud carrying the line throughout.
A ten-minute medley on the oud alone — no voice, just the strings.
A qaraami melody in the classical riftoon manner.
A patient qaraami performance built on the oud’s pulse.
Balwo verses over the oud — the short love-lyric form qaraami grew from.
Slow qaraami — the oud and the melody leaning on each other.
Radio Hargeisa & the archive
Born in Hargeisa with the 1943 invention of the Balwo and broadcast on Radio Hargeisa, classical Somali music lived on magnetic reels and cassettes. In 1988, station custodians heroically buried thousands of master tapes underground to save them from bombardment. Today, the hiss grows louder and magnetic tape decays — what was saved from the sands must now be written into score before it fades.
cassette-era masters · decaying
The work
The pipeline separates the voice from the band, finds the pentatonic root, scale degrees, and beat grid, then engraves the melody as sheet music you can play, print, or export. Where the model is less certain, it prints lighter ink instead of pretending — the score never claims more than the recording gives it.
The signature
Every note carries the confidence of its own transcription: full ink at certainty, fading ink where the tape fades. Uncertainty is stated on the page, the way a careful edition would state it.
the ink is the confidence
The archive
A listening room with every source stated plainly. A studio where the score, the waveform, and the recording share one cursor. A research corpus behind both — across AI, computational ethnomusicology, and cultural preservation.
Somali Kaban (Oud) · Cultural Heritage
Somali Music AI · Architecture & Capabilities
QaraamiGenAI combines generative machine learning, neural audio stem separation, and computational ethnomusicology to bring Somali musical traditions into the artificial intelligence era.
Generate authentic Somali Qaraami songs, solo Kaban improvisations, and traditional rhythmic pieces using fine-tuned neural models trained on archival recordings.
Convert old cassette tapes into accurate sheet music. Automatically detects Somali pentatonic modes, separates stems, and outputs playable PDF, MIDI, and MusicXML.
Compose in the browser with authentic virtual Somali instruments: Kaban (oud), Durbaan frame drums, and hand claps locked to Somali musical scales.
Everything you need to know about the first artificial intelligence platform dedicated to Somali music.
Somali Music AI (QaraamiGenAI) is the first artificial intelligence platform engineered specifically for Somali music and cultural preservation. Unlike generic AI models that default to Western scales, QaraamiGenAI is trained on traditional Somali pentatonic modes (5-note scales), Kaban (oud) acoustic ornamentation, Durbaan rhythms, and historic Radio Hargeisa recordings to generate, transcribe, and preserve Somali music.
The platform uses generative audio neural networks fine-tuned with custom LoRA adapters trained on hundreds of hours of authentic Somali recordings. You can generate authentic Somali genres—including Qaraami, Heello, Dhaanto, and Buraanbur—by specifying traditional scales, tempo (BPM), and instruments, producing audio that strictly honors traditional Somali modal structures.
Yes. QaraamiGenAI features a complete transcription pipeline that isolates vocals and instruments via neural stem separation, extracts pitches with deep CREPE models, fits the melody to authentic Somali pentatonic scales, and engraves playable sheet music. Transcriptions can be previewed with synchronous audio playback and exported to PDF, MIDI, or MusicXML.
The browser-based Somali Music Studio (DAW) includes virtual sampled Kaban (Somali Oud) with pentatonic scale lock, Durbaan frame drums with traditional groove presets, celebratory hand claps, acoustic bass, multitrack audio recording, and professional mixing tools built specifically for Somali compositions.
Founded in 1943 as the first Somali radio station in history, Radio Hargeisa was the cradle of modern Somali song and recorded masters like Cabdullaahi Qarshe, Cabdi Sinimo, Cumar Dhuule, and Maxamed Mooge. Many master tapes survived only because staff buried over 12,000 reels underground during the 1988 bombardment. Somali Music AI converts these fragile analog recordings into permanent digital notation and playable AI models.
Yes. QaraamiGenAI provides public access to AI-generated Qaraami recordings in the Listening Room, transcription examples, educational resources, and browser-based music creation tools for researchers, students, and musicians worldwide.
QaraamiGenAI was created by Khalid Ibrahim, an AI engineer and musician based in Minneapolis, MN. Combining modern artificial intelligence with traditional music training, Khalid created the platform to ensure Somali musical heritage is documented, understood, and revitalized for future generations.
605
recordings surveyed in the research corpus
105
cassette tracks digitised and on hand
412
notes in the first end-to-end transcription
0.68
its mean confidence — printed, not hidden
The world’s first Somali Music AI platform. Nothing here will ever be lost again.
End on a single held note