A Complete Guide to AI Music Generators in 2023
AI music generators have been popping up all over the place!
In fact, there's so many web apps and standalone programs today that the expression ai music is becoming a catchall phrase like "rock music". It's not specific enough to know what someone's talking about.
We need a better vocabulary for the sub-genres of software that are emerging. In this article, we've mapped out some of the trending AI music tools in December 2023.
Table of Contents
Section 1: Text to Music AI software
Section 2: AI voice generators
Section 3: AI DAWs for musicians
Section 4: AI audio-to-midi converters
Section 5: AI music generators for beginners
Text-to-Music software
Text-to-image apps like Dalle 2 and Midjourney have created a demand for similar experiences in song and video generation. As of 2023, there are a handful of AI music generators that take us closer to that goal.
AudioCipher: Text-to-MIDI VST
Website: AudioCipher
Cost: $29.99 one-time purchase (free lifetime version upgrades)
AI Music App Format: VST/AU plugin formats, Standalone option
Music output format: Midi
Target audience: Musicians who use a DAW
AudioCipher is a text-to-midi DAW plugin that uses musical cryptograms to turn phrases into chord progressions and melodies. Type in a word, choose your key signature, chord type, and rhythm automation.
Drag the words over into your MIDI instruments and apply the sound design that you have in mind. Refine the MIDI until you arrive at something you like.
The plugin currently uses a home brewed algorithm and does not run any AI services, however the team is working closely with machine learning engineers to develop prompt-to-midi services in a future version of the app. All sales cover operating costs, including research and development of an AI midi model.
Learn more about AudioCipher
Riffusion 2.0: AI Song Generator

Website: Riffusion
Cost: Free
AI Music App Format: Web app
Music output format: MP3
Target audience: Musicians and non-musicians
Riffusion was originally released in December 2022 as the first free AI text to music web application. Then, in October 2023, Riffusion announced that they had raised $4M in funding and rolled out a more advanced version of their website.
The app currently generates instrumental music in any style, with an optional second layer for vocals. Type in lyrics and Riffusion's AI will sing along with the instrumentals it creates. Store the audio in your user colelction and download the files as needed.
Chirp: Suno AI's lyric-to-song web app
Website: Suno.AI
Cost: Free, unlimited use
AI Music Format: Stream on Discord, download as MP4
Music output format: MP4
Target audience: Ai enthusiasts, lyricists
Chirp is a fun and easy-to-use platform that turns your lyrics into 20 second song sketches. Like Midjourney, the community operates on Discord and all of your music is created there.
Splash Music: AI beat maker with AI rapper

Website: Splash Music
Cost: $10/mo
AI Music App Format: Web app
Music output format: MP3 and mp4 (wav is on paid plan)
Target audience: Musicians and non-musicians
Splash music is a generative AI music company that specializes in electronic music. Free users can generate short instrumental audio clips and experiment with one of the AI lyric-to-singer voices. Build up a collection of content and download it at any time to use in your own personal projects.
Mubert Text-to-audio web app

Website: Mubert
Cost: Licensing structure ranges from $19-499 per track for a 45 second track
AI Music App Format: Web app
Music output format: MP3 and Wav
Target audience: Musicians and non-musicians
Mubert is currently the one of the best text-to-music solutions for generate full songs without unwanted artifacts. Their system takes in written descriptions along with qualities like mood. They pass it through an ML model trained on their labeled audio dataset. Existing human-made loops in the database are modified using composition algorithms and a layer of sound design, delivering a song that you can safely use in your podcasts and YouTube videos.

Google MusicLM

Website: Google MusicLM
Cost: Free
AI Music App Format: AI test kitchen (Google standalone app)
Music output format: Download feature available to select users only
Target audience: Musicians and non-musicians
MusicLM delivers the most accurate text-to-song generation service. Data scientists from Google's team measured the parity of MusicLM's audio output relative to the initial text and found it outperformed both Riffusion and Mubert. The audio quality is much better than Riffusion, but inferior to Mubert due to the presence of noise and artifacts.
For those who are interested, we've written an intro to Google's AI music datasets and how they trained the models that power MusicLM. As the owners of YouTube, Google has the right to train on any video in their collection. Their datasets include a human-labeled spreadsheet called MusicCaps, a medium-sized collection called AudioSet, and a massive 40+ million file, unsupervised dataset called MuLan.
WarpSound
Website: WarpSound
Cost: Unknown
AI Music App Format: API (for Twitch streams)
Music output format: Streamed audio on Twitch
Target audience: Gamers and live streamers
WarpSound is an adaptive AI music creator that lets users create music and publish NFTs on OpenSea. In May 2023 they announced an upcoming API that would support GPT-4 text-to-music generation for live youtube or twitch streams. They specialize in continuous music that adapts to incoming prompts. Like Mubert, their audio consists of an AI MIDI composition layer and another generative layer for the instruments themselves.
AI voice generators
Vocal deepfakes of mainstream artists went viral in 2023, with one AI Drake song Heart on My Sleeve generating over twenty million streams in under one month. AI music has become synonymous with AI songs, and with voice impersonation by extension. To learn more about specific products, check out our article on the best AI voice generators for music producers.
Vocal and Instrumental stem splitters

Stem splitting is commonly used to remove vocals from an existing track. Spotify does this as part of their AI karaoke feature, for example, so that users can sing along to the instrumental versions of their favorite songs.
For musicians, the most popular AI stem splitter is Lalal.ai due to its high quality output and API, which many other companies have white labeled and repackaged under their own brand. There are several other options available that you can read about here.
AI mixing and AI mastering software
Mixing and mastering tracks can be challenging, even if you're a seasoned musician. It used to require a trained ear and expensive hardware, which later gave way to more affordable DAW plugins. MusicRadar provides a complete list of smart plugins for mixing and mastering here. For a quick and simple solution, Landr offers a popular AI mastering tool that runs right in your web browser.
AI DAWs for musicians
The notion of an AI DAW is still relatively new. One founder coined the term generative audio workstation to describe the future set of tools that might be soon become available through artificial intelligence.
WavTool and Audio Design Desk are our personal favorites because of their innovative improvements to existing workflows. AIVA and BandLab are simple AI music generators that include a browser-DAW, which is why we include them in this category.
WavTool GPT-4 DAW
Website: WavTool
Cost: $20/month
App format: Web app
WavTool is the first AI DAW to offer a GPT-4 music production assistant. As the video above demonstrates, users can ask the copilot to create music tracks, add effects, and even compose MIDI. It excels at prescriptive instructions and struggles with abstract statements like "write a good song" or "write a catchy melody".
We've published a thought piece explaining how AutoGPT music prompts could improve on the limitations of call-and-response methods found in Wavtool or when using ChatGPT music prompts.
WavTool added a new sample generator function. One of our team members reviewed it in the clip below.
Audio Design Desk
Website: Audio design Desk
Cost: Three plans available
Basic: ($8.99/month or $89/year) 8 tracks and 50 regions, 3000 audio files
Personal: ($14.99/month or $149/year) Unlimited tracks, regions, 70k audio files
Professional: ($29.99/month or $299/year) Commercial licensing options
App format: Desktop DAW
Audio Design Desk is a sound design DAW that uses sonic intelligence to position and hot swap audio files relative to video. Users can trigger entire classes of audio, like risers and impacts, with the stroke of a key. This makes it easy to build full sonic landscapes for a scene in a single take.
ADD has won Product of the Year and Best of Show at audio conventions like NAMM and NAB for the past two years. The next version of Audio Design Desk will include an embedded suite of tools from their sister company, Makr.
You can demo the app directly on their homepage or watch a live presentation from one of their product experts below.
AIVA Song Generator

Website: AIVA
Popularity: Medium
Difficulty level: Low-to-medium
Music Genres: Modern Cinematic, Electronic, Ambient, Rock, Fantasy, Jazz, Sea Shanty, 20th Century Cinematic, Tango, Chinese
Generative Attributes: Mood, genre, theme, length, tempo, instruments
Cost: Three plans available
Free plan: Download up to 3 tracks per month in MIDI and MP3 format
Standard: (€15/month or €132/year) include 15 downloads per month
Pro: (€49/month or €396/year) includes 300 downloads and copyright
AIVA was one of the first companies to solve instant-music generation with AI. It runs in your internet browser, using deep learning algorithms trained on over 30,000 human compositions over several years. You don't need to play an instrument or know anything about making music to use it. AIVA does include a MIDI editor that requires more awareness of digital music arrangement.

Bandlab SongStarter

Website: Bandlab SongStarter
Popularity: Medium
Difficulty level: Low-to-medium
Music Genres: Pop, Trap, Hip Hop, Retro Wave, Acoustic Pop, Lo-fi, Electronic, Latin, Ambient, RnB
Generative Attributes: Genre, tempo, key signature, effects
Cost: $14.95/month or $149.50/year
BandLab is a popular browser DAW that offers an entry-level AI music creator. Users select their preferred type of music and then pick from one of three variations. BandLab makes the track's stems available within their digital studio. So while the generative attributes are limited, musicians can make changes to the tempo and key signature after the fact. Like most DAWs, you can control panning, volume and effects as well.

AI audio-to-midi converters
These audio-to-midi apps could come in handy for musicians that can't or don't want to decipher chords or melodies from recorded songs. Music producers who want to apply autotune and fine-tune their vocal performance will find Melodyne to be the most useful option here.
Samplab Polyphonic Audio-to-MIDI Converter
Website: Samplab
Cost: $9.99/month or $99.99/year
Polyphonic audio supported: yes
App format: DAW plugin and standalone
Samplab 2 is currently the leading polyphonic audio-to-MIDI converter. Users upload an audio file and select the region that they want to transcribe. The artificial intelligence layer detects separate instruments and converts each one into a separate MIDI track. In the past six month, the company has rolled out several updates and improvements, outlined in the video below.
Spotify Basic Pitch

Website: Basic Pitch
Cost: Free
Polyphonic audio supported: yes
App format: Web app
As one of many Spotify AI music apps, Basic Pitch is probably the most pragmatic creative tool for music producers. This polyphonic audio-to-midi converter loads directly within your browser.
Basic Pitch doesn't separate instruments the way Samplab does, and its accuracy hinges heavily on the source audio's complexity and noise levels. You can preview the MIDI in-browser and download the file directly from the site if you want to keep it.
Melodyne 5 by Celemony

Website: Melodyne
Cost: $399
Polyphonic audio supported: yes (only one instrument at a time)
App format: Plugin and standalone options
Melodyne 5 is generally considered the gold standard for vocal audio-to-MIDI conversion and editing. The high cost has put this software out of reach for your average audio engineer, but the transcriptions are highly accurate and the piano roll editor gives users unparalleled control over the articulation of each note. Melodyne does support polyphonic audio but only for one instrument at a time.
AI music generators without text prompts
To get started, we'll share a complete list of the most popular web apps that anyone can use to create music for commercial use. These music tracks don't run the risk of copyright strikes that you encounter by using commercial music without permission from the artist.
SOUNDRAW.io

Website: SOUNDRAW.IO
Popularity: High
Difficulty level: Low
AI generated music attributes: Mood, genre, theme, length, tempo, instruments
Cost: $19.99/month or $203.88/year
Users can explore existing tracks in the SOUNDRAW music database or pick your own attributes and hit create to start discovering unique tracks for your project.
A simplified AI music generator, called Ecrett, identifies itself as Soundraw's "little brother" and claims to have generated over 54 million combinations of music.

Boomy

Website: Boomy
Popularity: High
Difficulty level: Low
AI generated music attributes: Genres, Instruments, Drums, Mixing, SFX, Tempo
Cost: Three plans available
Free Plan: limited saves (undisclosed qty) and 5 album/single releases
Premium: ($2.99/month) Unlimited saves, downloads, 20 releases
Pro: ($9.99/month) Unlimited songs, releases, and priority rights requests
Boomy is an AI generated music community that lets users create original material and collect royalties through streaming platforms. The web app offers an extremely low barrier to entry, with just a few clicks required to create a song.
Over 14.5 million songs have been created with Boomy, representing 14% of the world's total recorded music. The app came under fire in early May 2023 when Spotify accused Boomy's creators of spamming the service with mobile farms that run up the streaming count.
Tens of thousands of songs were taken down and many sites incorrectly reported that these takedowns were due to Boomy's low quality music. Spotify clarified its position about the stream farms and reinstated the Boomy publishing pipeline a week later.

Soundful

Website: Soundful
Popularity: Medium
Difficulty level: Low
Music Genres: Lofi hip hop, Hype beats, Relaxation, Club beats, Trap, Dance, Drill, Afro pop, Disco, Pop funk, Synth pop, EDM, Techno, RnB, Latin, Reggaeton
AI generated music attributes: Genres, Instruments, Drums, Mixing, SFX, Tempo
Cost: Three plans available
Free Plan: 10 downloads, 1 stem pack
Premium: ($9.99/month or $89/year) Unlimited downloads, 5 stem packs
Pro: Cost is undisclosed, custom solution, monetize your songs
Soundful offers a more nuanced than Boomy in its approach to music generation. You can pick from a greater number of genres and music attributes, but it still has a simple enough interface for anyone to use. If you want to generate tracks for commercial use, you'll need to pay an additional fee.

Amper Music AI and Shutterstock

Amper was a powerful, browser-based AI music creator that delivered a similar feature set to AIVA, but with more controls over the final output. Shutterstock acquired Amper in 2020 and assimilated the app's musical output into a static music library. You might find it included in Twitter lists of "best AI tools" but technically speaking, it's just a stock music collection now. It's not going to create new songs but you can find royalty-free music for your corporate videos if you're trying to avoid copyright strikes.

MuseNet AI MIDI Generator

MuseNet was a browser-based AI MIDI generator app from OpenAI. When it launched in 2019, users could select from a few parameters like style, instruments, and length. The MIDI would playback in your browser and there was an option to export the MIDI file for free. An independent developer, Steven Waterman, built a UI layer on top of the MuseNet API called MuseTree.
Unfortunately, OpenAI switched off the MuseNet API near the end of December 2022, the same week that ChatGPT came out. The Musenet page now features a static demo that mimics the MIDI generation functionality. It always generates the same MIDI output. MuseTree no longer works, due to the MuseNet API being decommissioned.
This concludes our overview of AI music apps in 2023.
I hope you've discovered some new tools and that this list gave you a better sense of what's out there. Subscribe to our mailing list from the homepage to stay up to date on the latest in this field.