Interaction with the API of voice generators and sound trails

Desktop Apps — incorrectly specified categories?

290 USD

Project translated automatically. Log in or register, to view the original Everyone is greeted.There is a very interesting task, to create a voice from the text and to edit the same by structure audio routes by reference.Audio trails are created by a neural network by text, and the text is translated into different languages, for example - the original in Russian, and the text is also translated into English and Spanish, so dictors speak a little differently.It is necessary to create a software that will allow to do the following: Interacting with the external BD (Google table); Interact with the Voice Generator API (Microsoft, Google, Yandex and Zvukogram); - to regulate excess pauses at the source level; - to make a break between the segments at the source level; - to drive the audio trails of the imitators under the reference; to verify the correctness of the structure of the audio trails of the imitators; - stabilize the speed of the auditorium under the reference, according to the understandable script; - stabilize the speed of the audio segment of the series under the reference; - to receive processed (repared) audio path simulators in the required folder.Budget for the task: 13,000 rubles.Payment is made through this freelance site, the commission takes on itself.The detailed TZ can be read here: by request

Proposals 1