The Claim
An AdaBoost machine learning model trained on five speech pause features achieves 81% balanced accuracy in distinguishing cloned voices from authentic voices under controlled conditions with 49 participants and three cloning tools, and performs better than support vector machines and random forest algorithms.
What the research says
Not yet evaluated
We are still looking at what the research says.
These are independent scores, not a percentage. Higher-grade studies count more, so a single strong opposing study can outweigh several weaker ones.
A machine learning model using pauses in speech can correctly identify cloned voices 81% of the time in a controlled test with 49 people and three voice cloning tools, and it outperforms other common machine learning methods.
See the scientific wording
An AdaBoost machine learning model using five speech pause features achieves 81% balanced accuracy in distinguishing cloned from authentic voices in a controlled setting with 49 participants and three cloning tools, outperforming other classical algorithms such as SVM and random forest.
When a person speaks naturally, their brain controls the timing of pauses between words with precise muscle movements in the throat, tongue, and lungs. When a voice is cloned, the artificial system cannot perfectly copy these subtle timing patterns, so the pauses between words are slightly different. These differences show up in how long the pauses last and how they are spaced, allowing a computer to tell the real voice apart from the fake one.
What the research says
1 studyScientists used a computer program that looks at how long people pause when speaking to tell real voices apart from fake ones made by AI. This program got 81% of the answers right—better than other common computer methods tested.
Score breakdown, mechanism chain, raw evidence, ideal studies needed & 1 supporting studies
Not medical advice. For informational purposes only. Always consult a qualified healthcare professional before making health decisions.