"KOTODAMA" is a concept found in Japanese folk beliefs, where it is believed that words possess a certain power and meaning that can influence things or events. Users can input various types of text, such as blogs, textbooks, news articles, and more. Then, with the power of "KOTODAMA," the text will be transformed into specified styles, such as radio-style dialogues or comedy skits, and appropriate human voices will be added empowered by Eleven labs. As a result, even if the same text and mode are selected, you can enjoy different voices each time! The specific processes are as follows: First, the input text is converted by OpenAI in the specified style, and then the converted text is segmented at each speaker. Next, the ElevenLab API is used to convert the voices. Finally, the converted voices are combined and saved as an audio file. With these processes, our apps can give a lot of fun to mere text, thanks to the power of KOTODAMA.Category tags:
Walaa Nasr Elghitany
Data scientist and doctor
"What a creative and fun project! I really enjoyed listening to the different styles and voices that KOTODAMA can produce from any text. You have used OpenAI and ElevenLab APIs very effectively to transform text into engaging audio content. I think this project has a lot of potential applications in various markets, such as education, entertainment, and accessibility. One suggestion I have for improvement is to add some background music or sound effects to the audio output, depending on the style and mood of the text. This would make the audio more immersive and realistic, and also enhance the emotional impact of the words. Overall, I think you have done a fantastic job with the KOTODAMA project. Blessup for your innovation and skill! "
Crypto and NoCode Entrepreneur