This is a paid asset, but now you can download Minimal LipSync (Redux) for FREE, but please remember this package is provided only for learning purposes or to be able to test before buying the product, NOT FOR COMMERCIAL PURPOSES.
Detail this asset from Unity Store: Original Link
Minimal LipSync (Redux) v1.0
Speech based LipSync Implimentation using blend shapes
Minimal LipSync (Redux)
by Chippalrus
Speech based LipSync implimentation for Unity
Uses 4 blend shapes. Works with Microphone Input and AudioClips. Can work for 3D models. It isn’t perfect or very precise, but for cartoon style characters this works pretty well.
Only 2D Example is provided, but same method and setup can work for 3D models.
Uses four blend shapes to produce mouth shapes
Limited in precision, but functional for toon-like applications
Works for both Microphone Input and AudioClips
Can work for 2D and 3D
Uses Unity’s built-in Audio System, but could, in theory, work with FMOD or LASP ( Not tested for LASP or FMOD )