Method, device and system for real-time synthetic speech detection in a resource-constrained environment

A synthesis speech detection system, the system comprising:

• a synthesis speech detecting device;
• and
• a synthesis speech detection model training device,
• wherein the synthesis speech detecting device includes:
• a first processor;
• a first display;
• and
• a first memory including a pre-trained synthesis speech detection model,
• wherein the first processor is configured to:
• generate synthesis speech detection data of input speech data by using the pre-trained synthesis speech detection model;
• and
• allow the generated synthesis speech detection data to be displayed through the first display,
• wherein
• for detecting synthesized speech in real time on one device which has computing resources less than a predetermined threshold,
• the generating of the synthesis speech detection data by the first processor includes:
• generating pieces of input speech section data by dividing the input speech data in a sliding window method;
• generating pieces of section synthesis detection data for the pieces of input speech section data, respectively;
• and
• generating synthesis speech detection data of the input speech data by using a movement average of the generated pieces of section synthesis detection data,
• the synthesis speech detection model training device includes:
• a second processor;
• and
• a second memory including a teacher model and a student model,
• the second processor is configured to:
• generate the student model initialized by using at least part of one or more teacher model layers included in the teacher model;
• input learning speech data into the teacher model and the student model;
• generate teacher extraction data from the teacher model and student extraction data from the student model;
• and
• train the student model based on the teacher extraction data and the student extraction data,
• the second processor is further configured to:
• initialize a weight of at least one student model layer included in the student model to a weight of at least one corresponding teacher model layer,
• acoustic features are extracted from layers forming a part of a front end of the teacher model,
• the student model for processing the acoustic features is constructed by initializing the part of the front end layer of the teacher model to the student model,
• the student model has a number of layers fewer than the teacher model,
• the number of student model layers included in the student model is smaller than or equal to ¼ of a number of teacher model layers included in the teacher model, and
• the pre-trained synthesis speech detection model is processed in an end-to-end manner within the synthesis speech detecting device composing of a single device.