Loading request...
The user was unsatisfied with CassetteAi's music generation because specific instructions for adding a phrase as vocals and a deer call were ignored. This indicates a need for more precise control over the inclusion and generation of specific vocal phrases and sound effects.
Day 17 of #100daysofai: Generate Music and another Video Today's first task was to create a few seconds of music with generative AI. The platform was @CassetteAi, which provides a simple workflow to get started. You have to input text like in ChatGPT to describe the music which should be generated, and you will get a creation in less than a minute. To make it even easier, you have multiple presets you can combine and create out of these a prompt with the "refine" button. You can also upload your own audio file to guide the generation process and add various tracks from multiple generations. However, my first tries were not really satisfying as my instructions of adding a specific phrase as vocals as well as the addition of a deer call were ignored. The second task of today was to create a video, like done yesterday with Pika, but this time with a service called Genmo. I used the same image as yesterday as input. However, the result is by far less stunning as the generation from Pika shared in my post from Day 16. One year back, the result from Genmo would have been impressive, but the AI video generation improved a lot in the last couple of months.