Nowadays, Twitter is largely used to access or share information related to a large variety of topics. In recent years, researchers have shown that Twitter data are very helpful in several fields and may have many applications. In June 2020, Twitter has announced voice tweets to create a more human experience. This paper gives an overview of possible consequences of this new feature. Furthermore, we present some challenges that researchers will face to exploit this type of data. Finally, we also introduce several opportunities for future research related to voice tweets.

In past years several works have noted that Twitter data are essential in diverse fields and may have a lot of applications. Nevertheless, the API proposed by Twitter sternly restricts access to public data generated by users. These restrictions have the consequences of greatly slowing down the contributions of researchers and of limiting their scope. In this paper we introduce TwiScraper, a collaborative project to enhance Twitter data collection by scraping methods. We present a module allowing user-centered data collection: Twi-FFN.