Table of Contents[Hide][Show]
- How Does Riffusion Work?
- How To Use Riffusion
- How The New YouTube AI Tools Can Help You Create More Engaging And Creative Videos
- How To Manually Identify AI-Generated Text Without Using A ChatGPT Detector
- AI Lawyers? Not So Fast! Survey Explains Why Most Lawyers Aren’t Using AI… Yet.
- Did A Bot Write That? 5 Free AI Detection Tools That Can Help Expose ChatGPT Cheating
- Giant List Of Writing Styles To Use With ChatGPT And Google Bard
- 15 Useful ChatGPT Prompts For Sales
- How To Make A Presentation In 60 Seconds With Gamma.app: The AI PowerPoint Generator
- Did A Bot Say That? – ElevenLabs Says Their AI Speech Classifier Is 99% Accurate
- Castmagic Review: The AI App That’s A ChatGPT For Podcasters And YouTubers
- Upgrade Your ChatGPT Prompts With These 12 Smart Engineering Tricks
- Only 14% of U.S. Adults Have Tried ChatGPT. Have You?
- The 11 Biggest ChatGPT Flaws And Why You Should Never Completely Trust Online Chatbots
Riffusion is a new application of AI technology for the music industry. Developed by Seth Forsgren and Hayk Martiros as a hobby project, it utilizes the same model as the widely recognized AI tool ChatGPT to generate unique musical compositions from text prompts.
How Does Riffusion Work?
What’s the secret behind its magic? Spectrograms! Riffusion utilizes AI to convert spectrograms, images generated from audio, into unique audio tracks. By varying the starting point, or “seed,” an infinite number of variations can be created from the same text prompt.
The technology behind this process involves inverting the Short-time Fourier transform to reconstruct audio from spectrograms. The AI model used by Riffusion only considers the amplitude of the sine waves and the Griffin-Lim algorithm is applied to approximate the phases during audio reconstruction.
NERD NOTE: The Griffin-Lim algorithm is a method used to reconstruct audio signals from their magnitude spectrograms, using an iterative process to minimize the difference between the original and the reconstructed signals. It was invented by Daniel P. W. Griffin and Jae S. Lim in 2010, and it is used to approximate the phases during audio reconstruction from Spectrograms.
How To Use Riffusion
The Riffusion web application allows users to input text prompts and generates interpolated content in real-time, providing a visual 3D representation of the spectrogram. The app also allows users to quickly move on to the next prompt, and if none is provided, the music AI will interpolate between different seeds of the same prompt.
Here are some examples:
Once you create a sound loop that you like, simply click on the share option to create a sharable link that you can send others.
What’s Next For AI Music Generation?
Riffusion is a noteworthy addition to the field of AI-generated music, and it has the potential to further enhance the creative process for professionals in the industry. While the inner workings of its AI generated music technology may be complex, the results are impressive and showcase the potential for future advancements in the field of AI music generation.
Whether you’re a songwriter, composer, or music producer, AI music generators are a great music brainstorming tool for creating endless variations of your ideas.
You can experience the capabilities of Riffusion for yourself by visiting their website at riffusion.com.