Inspiration --- Wanted to have an easy way to make a freestyle song for any occasion, without the hassle of thinking about the whole lyrics and even the music selection!
What it does --- It takes a user to a web link which asks the user for his/her likes, dislikes, place, friends, etc. These entities are then included in the song on that website. The user can then choose one of the three instrumental tracks that are available in the app. Once the user selects the track, he/she can press on "Play" button and the track will play along with the text-to-speech version of the song. There is an option for adjusting the speed of Text-to-Speech so as to synchronize it to the music.
How I built it
I built it on Android platform.
Challenges I ran into
I could not find a way to copy the actual song from the webpage to the text-box directly. Also, there is no free service for text-to-speech which allows anyone's voice yet.
Accomplishments that I'm proud of
The inception of a good idea after scraping the first one off at 3PM.
What I learned
I learned that new API, limitations of some API, limitation of research or even implementation in the Text-to-Speech area.
What's next for FreeStyle
Many languages are available for Google Text-to-Speech, so giving the option of making the song in multiple languages would be cool. The Tempo of a music track can be found out using this free code: https://github.com/javonharper/Tempo. This will allow the rate of Text-to-Speech to be calculated for synchronization with the song automatically. The voice is digital right now but can be converted to any human's voice by using "https://www.mivoq.it/en/" but it is a paid API. It would be interesting to try to convince the authors of that API to test out a song with their API for free. Of course, the UI is very basic, so it can be improved. Also, more options can be shown to the user, so that the music can automatically be selected from the large selection of music.