top of page
Search

WavTool: The World's First Text-to-Music AI DAW Uses GPT-4

We've been through a year of what feels like warp-speed developments in AI music software. The history of these text-to-music apps can be traced back to a pivotal moment in April 2023, when WavTool's AI DAW debuted with a GPT-4 chat assistant. It was able to help compose midi, generate instruments, control effects, and run activities in the workstation through AI chat commands.


WavTool announced a new round of features in January 2024 that got us excited to revisit the app and revamp this article. If you gave the app a spin during those early months, I encourage you to take another look. There's lots of new goodies to discover!



As of 2024, WavTool is the only AI DAW with a MIDI composer assistant powered by GPT-4. So let's get into it and see exactly what this tool has to offer!


Table of Contents


What is WavTool?



WavTool is an AI-powered digital audio workstation that loads in your browser. That's right, you never have to download an app, and with their one-click sign up through Google or Facebook, it's very easy to get started using the latest version of Google Chrome browser.


It includes an embedded AI chatbot, called the Conductor, that you can show and hide at will. The GPT-4 powered assistant can detect chords and understands audio production techniques. In this way, it acts as a kind of AI bandmate.


Here are some of the core features that came with WavTool back when we first reviewed it back in May 2023. They're still available today:


  1. AI Conductor - WavTool calls their AI chatbot the Conductor because of its ability to guide you through the music making process. It can touch every part of the DAW, hold deep conversations about music, and generate MIDI.

  2. Browser app with cloud saving - WavTool projects are saved to the cloud, so you can open your projects up on any computer after logging in.

  3. Custom Wavetables - Pull in one of WavTool's instrument presets for your MIDI track or build custom wavetables from scratch. The video above goes into detail on how users can synthesize new instruments.

  4. Device panel - WavTool's panel lets you set up devices that control EQ, reverb, delay, dynamics, distortion, LFOs and side chain compression, visualizers and more. Unlike Ableton Max (a comparable tool), AI Conductor can create and edit devices from your text prompts.

  5. QWERTY & MIDI Controller Support - WavTool's piano roll includes a keyboard interface that shows you the notes you're playing. You can play on a QWERTY computer keyboard or use a standard MIDI controller.


What's new at WavTool?



Many of our readers will be familiar with WavTool's classic features listed above, but might not have kept up with more recent developments. They've maintained an impressive monthly cadence with several new features released. Let's have a look at the timeline below and see what their Library has to offer.


Wavtool Library feature

Wavtool rolled out a new Skills interface in October 2023 that included a number of special AI features and workflows:

  • Generate short loops: Turn your musical ideas into five-second samples using WavTool's text-to-music generator

  • Humanize Clip: Quickly add rhythm and velocity variation

  • Convert to MIDI: Convert audio to MIDI with controls to fine-tune the conversion in real time

  • Stem Splitter - Audio clips can now be split into vocals, drums, bass, and any remaining sound. This makes it easier to remove background noise from rough recordings and create a mixable, multitrack project. It's more convenient than toggling between their app and other AI stem splitters.


Exporting and sharing songs

Song Sharing URLs were introduced in November 2023, making it possible to host your music on a public URL and share with others. Just select the song section you want to share and click Export -> To Song Page. Name the track, generate a link, and you're good to go.


WavTool prototypes

Prototypes were introduced in December 2023, delivering a popular AI music feature called Timbre Transfer that captures the sound of one instrument and applies it to any audio clip. They offer a handful of output sounds to experiment with as well as the option to train your own sound models.


Creating custom voice and instrument models

Some major updates came out in January this year, including improvements on existing skills and a DAW bridge that lets you load your plugins into the browser app!


  • Voice Model (Prototype) Training - Users can now create their own custom Prototypes (train an AI model) for any consenting voice, and any non-voice sound.

    • Non-voice models have all vocal sounds removed from them prior to training, and voice models are only trained if the words spoken in the training data substantially match a time-limited unique AI-generated training script.

    • Due to the verification requirement, it should not be possible to train a voice model from pre-existing recordings of a person speaking or singing.

  • Skills Refresh - Skills are now included in the Library panel, and can be opened simultaneously, allowing for a seamless and dynamic workflow. Drag and drop skill inputs and outputs into a project or between skills.

  • Conductor Upgrade - Conductor can now chain together multiple AI processing steps/Skills. For example, the prompt “Make some disco music, convert the bass line into MIDI, and add some alternate chords to it” triggers the following Skills:

    • AI text-to-audio sample generation for ‘Disco Music’

    • Stem Split to pull out the bass line

    • Audio-to-MIDI conversion of the bass line

    • Composer suggests a new set of MIDI notes above the bass line


These are great, but as a text-to-MIDI plugin company ourselves, we were particularly giddy to find out that WavTool now offers Plugin Support. This is a massive step forward for them and legitimizes them as a DAW.


How to load any audio VST in WavTool


To get started, you will need to install the WavTool Bridge, a small plugin host that allows one to connect their installed plugins. There's a link to it in the product, but you can also access the plugin bridge page directly to download it.


  1. Install the plugin bridge and boot it up. You'll see the WavTool icon on your computer's top navigation menu, next to any other open applications.

  2. Open the Library panel in WavTool and select the Plugins & Devices tab.

  3. Scan your existing library and the plugin bridge will go through your local computer to find all of the plugins

  4. Once the scan is complete, double click on the plugin you want to load


In the example below, we've loaded the AudioCipher text-to-midi plugin. After typing in a test phrase and dragging the MIDI into the arrangement view, there was a short loading period of a few seconds. Just moments later, the MIDI notes were there as expected!


AudioCipher with WavTool DAW bridge

The DAW bridge file is only a few megabytes large and we were able to install it without any difficulties using a Macbook pro. We spoke to WavTool's team and they confirmed that Windows support is coming soon.


Each plugin runs in a sandboxed environment, meaning errors in the plugin do not affect the rest of the user’s session. This is a big perk, as many of the most advanced DAWs have been known to crash if a plugin hits a critical bug.


Conductor chatbot: AI music composer


WavTool's AI music composer, the Conductor, leverages the conversational intelligence of GPT-4 to have deep and nuanced conversations with you about any musical topic. But its real talent is the ability to turn around and take action in the DAW, based on commands from GPT. This text-to-music feature is something we've never seen before.


WavTool conductor

There's one very exciting thing about Conductor that sets the AI DAW apart from the other major AI MIDI generator apps. Unlike Google and OpenAI's MIDI generators, WavTool knows why it generated MIDI in a particular way and explain its reasoning to you in detail. You just have to ask!


Previous MIDI generators have not been set up to engage conversationally or take text commands. This means we never knew why the AI was generating a particular MIDI melody or chord progression. It was not possible to critique and fine tune its choices over time, other than requesting variations.


Prompting tips for the WavTool MIDI generator

MIDI generation with the conductor

As a WavTool user, you don't need be an amazing writer or "prompt engineer" to get started. That being said, the words you use with the AI will dictate the quality of its creations and the quality of your experience.


A fair warning for any music theory heads out there - WavTool's conductor cannot currently turn requests for a specific chord progression or melodic shape into MIDI. Instead, try to keep your requests limited to more general moods and styles. This is the best way to take advantage of the knowledge base that it does have and get the best results.


Pricing: How much does WavTool cost?


WavTool previously had a Free tier with a limited number of AI prompts, which posed restrictions on the length and number of tracks in your production. They have since revamped their pricing to allow all users to access all features expected of a DAW for free. They also have a free 2-week trial for the Pro tier.


WavTool pricing page

Final Thoughts on WavTool & the future of AI DAWs


Generative audio workstations and AI DAWs are two different names for the same basic idea. They stand in contrast to what could be called AI VSTs, or third party plugin / standalone apps that run AI services to augment a DAW.


There are a few other AI-powered audio workstations, like Output's Co-Producer that debuted in December 2023, the audio-to-midi RipX DAW, and a still unreleased Deepmind Lyria program by Google.


In this new landscape, our musical vocabulary will be an asset. Writers may gain a competitive advantage in music that they've never enjoyed before.


Melody generators and chord progression software could also be in the line of fire, if these AI DAWs become sufficiently advanced. Why would you pay for a random note generator when your AI music composer can do it for you?


That being said, GPT-4 still has some ways to go before it actually poses a threat to MIDI generation software. Producers are attached to their existing DAWs and workflows. The quality of GPT's musical output also needs to get better in order to claim the throne.


Sign up for WavTool here: https://wavtool.com/

Comments


bottom of page