I already have the 3D model of the character (boy) with the 3ds Max with voice-o-matic plugin project that generate about 6 minutes of the character talking (reciting) with very minimal hand and eye movement. This project if it runs it will generate the character reciting for 6 minutes.
The task to do is to take the project, use the character model in the first project to generate several animation segments like the original project where the character will recite several one to 3 minutes segments. It is all reciting where there are eye and head movements and very few hands movements. The total time of the generated animation will be one hour. I will give you the sound recordings for each segment in .mp3 and .docx of the transliterated English text of what the character is saying written in English alphabit. Exactly like what is in the first project. The lip synching is the most important thing. A low quality preview example is attached that was generated from the project of the first 6 min recitation.