AI tools for musicians: our selection (26)

AI for musicians

For musicians, AI can support idea development, rehearsal preparation and work on existing recordings. It can contribute at different stages without ensuring that the result fits your artistic intentions. AIVA can compose in different styles and use an audio or MIDI file as inspiration. LALAL.AI can separate vocals from accompaniment or isolate individual instruments. During a working day, you might listen to an isolated part before rehearsal, then explore a musical variation before returning to the arrangement yourself. Stable Audio can replace a section without regenerating the whole piece. These outputs remain working materials that need your judgement. Separation may introduce audible artefacts, and generated passages may contain transitions that do not suit the song. Interpretation, the choice of takes and communication with other musicians still depend on attentive listening and the decisions you make throughout a recording or rehearsal session.

Tasks where AI saves time

• Separate vocals from accompaniment to prepare a practice session using a recording you have permission to process.
• Isolate an instrument within a recording to hear its phrasing more clearly before rehearsing with other musicians.
• Reduce recording noise to prepare a working reference, then listen for processing damage and lost musical detail.
• Generate a song sketch from your own lyrics to explore a direction before developing the arrangement yourself.
• Replace a musical section within an audio excerpt to test a transition without regenerating the entire piece.
• Change the timbre of an authorised singing voice to discuss a different sound with the collaborators involved.

Our picks

Picks set by simple rules on the verified data, with no partnership:

• Best scored : Suno. 76/100 on our access and transparency grid.
• Cheapest paid option : ecrett music. First plan at $4.99 per month.
• Open source : TextFX. Public code, self-hosting possible depending on the project.

The best-scored tools at a glance

Suno 76/100

Suno is for people creating songs from descriptions, their own lyrics, or audio excerpts. It also supports separating a track’s elements and editing songs in a web studio. Features: Create songs from a description; Produce a song with your own lyrics; Turn an audio excerpt into a new creation. Entry price: $8 per month, with a free tier. Runs on: web browser, iPhone and iPad, Android, and API.

LALAL.AI 73/100

LALAL.AI helps people work with audio recordings. It separates vocals and accompaniment, isolates instruments, reduces noise, removes echo and reverberation, and lets users transform or clone a voice. Features: Separate vocals and accompaniment; Isolate different instruments; Reduce noise in a recording. Entry price: $7.5 per month, with a free tier. Runs on: web browser, Windows, macOS, Linux, iPhone and iPad, Android, and API.

PrismPoster 73/100

PrismPoster lets people creating multimedia content generate videos from photos or text, edit images, and compose songs with lyrics. It assembles media and reuses characters across scenes. Features: Create a video from a photo or text; Generate and edit images; Compose songs with lyrics. Entry price: $3.36 pay as you go, with a free tier. Runs on: web browser and API. Data hosting: Several regions.

AIVA 71/100

AIVA composes tracks in different styles and supports creating a custom musical style. It serves people who want to edit generated tracks or use audio and MIDI files as inspiration. Features: Compose tracks in different styles; Create a custom musical style; Use an audio or MIDI file as inspiration. Entry price: €11 per month, with a free tier. Runs on: web browser, Windows, macOS, and Linux.

Voice-Swap 68/100

Voice-Swap serves people working with singing or voiceovers. It changes vocal timbre, creates voice models to share with collaborators and generates voiceovers with adjustable pacing and pauses. Features: Change the timbre of a singing voice; Create a model of your own voice; Share a voice model with collaborators. Entry price: £6.99 per month, with a free tier. Runs on: web browser and API.

Stable Audio 68/100

Stable Audio serves people creating music and sound effects from text or existing audio. It replaces sections, generates separate tracks, and lets users adjust mix levels and effects. Features: Create music and sound effects from text; Transform an existing audio excerpt; Replace a section without recreating the entire track. Free tier available. Runs on: web browser and API. Data hosting: United States.

In figures

Of the 26 tools selected for this job, 25 (96%) offer a free tier, 62% have a public API and 6 have a French interface. The median entry price, per month, is $9 (across 14 tools with a published price). 5 disclose where data is hosted, 0 of them in Europe or on device.

AI video generators

To create videos from text or images: 2 tools. See the whole category.

Voice cloning and changing

To clone or change a voice: 1 tool. See the whole category.

Creative writing and scripts

To write fiction, scripts and stories: 1 tool. See the whole category.

Job-specific precautions

Check your permissions for recordings, lyrics and reference files before uploading them. Obtain explicit agreement before transforming someone's voice or sharing a voice model, and define the permitted uses. Review the service terms governing distribution and use of generated material before including it in a project. Do not assume an output comes without rights restrictions. Listen carefully for artefacts, altered breaths and awkward transitions on suitable equipment. Protect unreleased demos and keep the source files, relevant service terms and agreements made with the other contributors.

Where to start

Choose a specific task, such as isolating an instrument before rehearsal. Try LALAL.AI with a recording you have the right to process. Compare the time needed to produce a useful practice reference with your usual approach. Include exporting, listening and any corrective work. Note passages where separation artefacts interfere with musical understanding. Keep the tool if it makes preparation easier in practice. If the isolated part remains unclear, try a different source or return to the original recording.

Want to go beyond a single tool? AI agents chain several tasks on their own; our quiz helps you pick one.