Voiceprint recognition, also known as speaker recognition, is a technique for discriminating the identity of a speaker by voice.
The voiceprint recognition principle converts the sound signal into an electrical signal, which is then recognized by a computer.
Process voiceprint acquisition, feature extraction, voiceprint modeling, voiceprint contrast judgment
Advantages Voice extraction is easy, speech recognition is low in cost, suitable for remote identification, recognition and confirmation algorithm is low in complexity
Disadvantages People's voices are susceptible to physical conditions, age, mood, etc., and are highly influenced by the environment. The voiceprint features of many people are difficult to extract.
Summarizing the requirements of voiceprint recognition data is actually much higher than speech recognition. There are many good application scenarios, but it is relatively affected by other factors. Overcoming these unfavorable factors, voiceprint recognition will have a wider application space.