The Melodrive difference


15+ years of experience researching and developing AI and machine learning solutions in the audio/music space. All our team members have a unique double background in AI and music.

Built for audio startups

Fast iteration cycles, pivots, business model canvas and lean methodology. We know how audio-centric startups work inside out and can fit into their workflow.

Long-term partnership

We’re focused on working with a few new clients each year where we believe we can have the most impact on their growth. We stick with our clients from the design of the AI strategy to ongoing R&D.

Your transformation

Learn how AI can help your startup

Build an AI prototype

Run ongoing

Develop an AI strategy

Deploy an AI solution

Getting from where you're today to the point where you’ve unlocked the power of your audio data with AI is a transformation that usually follows a standard path.

Once you understand how AI can help your business, it’s time to define a well-rounded AI strategy that can leverage metadata and audio content. Following the lean startup methodology, you should focus on implementing the most valuable AI solution first. As such, you can build a prototype and evaluate its results. If the prototype performs well, you want to iterate, upgrading the solution, and deploying it into production. If the feedback from your users is positive, you’re in business! It’s time to start ongoing R&D. The aim is to output new AI solutions that will provide an unfair advantage to the startup and fully leverage the potential of your data.


Our services consist of a comprehensive, integrated approach that supports audio-centric startups along their path to unlock the value of their data. Each service responds to a different need in your transformation process.

1:1 coaching sessions

You’re aware you’re sitting on a gold mine: your metadata and audio content. You know that the solution to unlock the value of your data is AI and machine learning. However, you don’t really understand what AI is, nor how you could implement AI solutions in your startup. You want to learn more about the world of AI audio and see what opportunities it may open for your startup.

If that’s the case, 1:1 coaching is ideal for you. In these sessions, we can:

  • help you understand how AI works
  • help put together an effective AI strategy
  • identify areas where AI could be beneficial in your business
  • advise on how to run an AI project at its best
  • advice on how to prepare your data for machine learning
  • help avoid pitfalls and rely on best practices when working on AI audio
  • help avoid pitfalls and rely on best practices when working on AI audio
  • help put together a world-class AI audio team
  • advise on how to choose the right employees to implement your AI strategy

1:1 coaching sessions are recurrent (weekly or every other week), with a minimum commitment of 1 month.

The fee for a 1-hour session is $300.

The initial touchpoint is a free session. After the first meeting, we’ll decide if we want to continue the relationship. Payment begins after that decision.

On-site workshops

You want to bring up a big change in your audio startup, by embracing AI technologies. However, your technical and non-technical employees are not familiar with AI. You need a crash course to get you and your team started with AI in the audio space.

In this case, the on-site workshop is the right service for you. We’ve developed workshops for both technical and business-oriented audiences, which introduce the attendees to the engineering and business/product aspects of AI applied to audio.

For the engineering teams, we focus on a hands-on approach in the Python programming language. In the session, we cover the following themes:

  • What’s audio digital signal processing?
  • What’s deep learning?
  • Tensorflow and Keras
  • Multilayer perceptron
  • Convolutional neural networks
  • Recurrent neural networks
  • Designing and developing an AI audio solution
  • Deploying an AI audio solution with Flask and Docker
  • Best practices and pitfalls for startups
  • Pointers to additional learning resources

The non-technical workshop provides a business understanding of AI audio technologies to co-founders and non-engineering teams. Attendees can use this knowledge to decide how and where to apply AI effectively in their startups. The areas we cover are:

  • What’s AI, machine learning and deep learning?
  • AI applications in audio
  • AI capabilities and limitations in audio
  • Industry case studies in AI audio
  • Identifying company areas where to apply AI
  • Creating an effective AI strategy
  • Planning an AI project
  • Managing an AI project
  • Building an AI audio team
  • Calculating the market value and ROI for an AI audio project
  • Running continuous R&D in AI audio
  • Additional learning resources

A workshop lasts for 3 days. The fee for a workshop is $14,000.

AI audio strategy consulting

You have a decent grasp of what AI is and how it could be beneficial to your startup. You plan to use machine learning to leverage your data. However, you don’t know how to put together a coherent AI strategy. You realise that AI audio is complex and that you need external help to organise a plan that may bring the most value out of your data.

If you’re in this situation, the AI audio strategy consulting service is what you should go for. We bundle the following approaches to arrive at a comprehensive AI plan:

  • Study the UVP of the company and its offering
  • Familiarise with the company talent
  • Study the company tech stack
  • Examine the data
  • Interviews with customers
  • Perform research to determine viable AI audio applications

Price starts at $30,000 and depends on the complexity of the project. The turnaround is 1.5 months or shorter.

We convey our findings through:

  • Weekly 30’ update calls
  • A final written report with the AI strategy. The report outlines viable AI applications which leverage your data, prioritised by value, risk and complexity. It also provides a step-by-step plan to implement the strategy and details the resources and changes that need to happen to implement the plan.
  • (Optional) A follow-on knowledge transfer engagement where we meet weekly or every other week to help you implement the AI strategy.

AI audio prototype

You want to build an AI audio application that will help grow your startup, give you a strategic edge over the competition, or increase the loyalty of your users. You think the idea is great, but you want to test it before launching it. You don’t have an internal AI audio team, so you can’t develop the solution in-house.

If you find yourself in this scenario, our AI audio prototype is the perfect choice for you. With this service we:

  • research the academic literature in AI audio to implement the best solutions
  • run in-house research
  • develop several concurrent solutions to test what works best
  • evaluate the solution

We use the Python programming language to develop the prototype.

The fee depends on the complexity of the problem but starts at $60,000. The turnaround is at least 2 months.

Production-ready AI audio solution

You have a precise AI use case you want to implement that will enable you to take advantage of your data. You need an end-to-end solution that takes your idea from research all the way up to deployment and, potentially, maintenance. You don’t have AI music talent in your startup. You’ve looked around to outsource implementation, but haven’t found an AI shop that works for you. None of these agencies seem to have a focus on audio, nor they talk the startup lingo.

If this is your situation, then consider going for the production-ready AI solution service. With this service we:

  • research the academic literature in AI audio to bring you the best solutions
  • run in-house research
  • develop the AI audio solution
  • evaluate the application
  • make the solution production-ready
  • (optional) deploy the solution as a Flask API
  • package the solution with Docker
  • setup the cloud infrastructure to serve the solution
  • deploy the solution on the cloud
  • update and maintain solution

We work with the Python programming language to implement AI audio solutions.

The fee is custom and starts at $120,000. Turnaround is at least 4 months. Pricing for maintenance and support starts at $10,000/mo and depends on the complexity of the project.

Ongoing R&D and support

You've already implemented an AI audio solution and have experienced its benefits. You're ready to go all-in on AI audio to leverage your data at their fullest potential. By researching and developing new AI audio solutions, you count to improve your offering to your customers and solidify your strategic advantage. If you rely heavily on AI, you know you'll get a better chance at raising the next round of VC funding.

However, putting together an R&D team with a focus on AI audio is difficult. Hiring AI talent is tough, finding AI audio talent even more so. You also don't have the expertise to run an AI audio team in-house and don't feel confident planning an AI strategy long term.

If this is your situation, then the ongoing R&D and support service is the right solution for you. This service establishes a long-term relationship between Melodrive and your startup. We become your R&D team. As part of the service we:

  • identify key areas where AI can give you a strategic edge
  • create an AI strategy and iterate on it
  • advice on promising AI directions to follow
  • research new AI audio solutions
  • develop new AI audio solutions
  • evaluate solutions
  • deploy solutions
  • maintain and update solutions

As part of this service, we use a lean startup approach, running R&D in agile sprints to match our outputs with the ever-changing needs of your startup.

The ongoing R&D and support package is very involved for both parties. We only work with startups we’ve established a great relationship and whose products/services align with our vision and values.

The ongoing R&D and support service is paid monthly, with a minimum commitment of 6 months. Pricing is custom and starts at $40,000/mo.

Learn more

We created a 6-part email course that speaks to many of the problems you’ll face as you try to unlock value from your audio data. The course presents a high-level roadmap you can follow to integrate AI audio technology in your startup.

You can unsubscribe at any time by clicking the link in the footer of our emails.

Knowledge centre

At Melodrive, we're on a mission to educate our audience about AI and its applications in the audio space. As such, we regularly publish research, interviews, tutorials and insights in different media formats, targeted to both technical and non-technical folks.

The sound of AI - Blog

In this Medium publication, we share research and the output of our activity in the AI music space.

The sound of AI - Business bites

Startup founders and executives can watch these videos to gain an understanding of AI and its applications in audio, from a business perspective.

Meet our team

Valerio Velardo, PhD


Valerio is an AI and music researcher, composer, conductor, pianist and programmer. He earned a PhD in Music and AI from the University of Huddersfield. He developed mathematical and computational models to simulate the behaviour of music societies. Valerio also studied astrophysics, with a focus on complex systems and its applications in the social sciences.

Andy Elmsley, PhD


Andy is a machine learning expert, games developer, artist and musician and with a keen interest in human/AI interaction and co-creation. He earned his PhD from City University of London where he focussed on real-time expressive rhythm generation using deep learning. He also once conducted a robot orchestra playing the Dr Who theme tune for the BBC.

Cárthach Ó Nuanáin, PhD

AI music programmer

Cárthach is a musician, composer and engineer with over a decade’s experience in the industry, arts and academia. He holds a Ph.D. in Computer Music from Universitat Pompeu Fabra and an M.Phil. in Music & Media from Trinity College Dublin. He has created many prize-winning systems for algorithmic and assisted composition, and in a past life produced audio for numerous Triple A game titles with Microsoft Games Studios.

Christian Tronhjem

Product and sound designer

Christian heads up our product team, and is a specialist in agile user-driven product development. He also holds a masters in Electronic Music Composition from the Royal Academy of Music, Denmark, with a special focus on game audio from DADIU. He’s worked as a composer, sound designer and sound engineer across different media; from award-nominated dance performances, short films, games and TV commercials, to mixing, mastering and producing death metal, jazz and everything in between.

Connect with Melodrive