Artificial Intelligence (AI) has made its mark in the modern world. For many years, humans have dreamt of a world with autonomous transportation and personal virtual assistants like Cortana, Siri, Alexa and so on. Almost 60 years later, autonomous vehicles and personal virtual assistants have become one of the most commonly used forms of AI technology. It is now part of an average human’s day to day life. AI can now do a lot more than facial recognition and voice recognition. It can bring to life portraits with just a single image.
Recent innovation in AI technology enables it to guess the appearance of a person with just their voice. Researchers at Carnegie University managed to stun people with their new AI experiment where a short audio clip of a person’s voice is played and the technology reconstructs what the person might look like in real life. The results were not completely accurate but they did provide a remarkable resemblance to what the person looks like in real life. The team described how they trained a generative adversarial network to analyze short voice clips and “match several biometric characteristics of the speaker,” resulting in “matching accuracies that are much better than chance.” Despite the Speech2Face algorithm being a cutting-edge invention, it also raises concerns with regard to privacy breach and discrimination. In a report, the team said, “Although this is a purely academic investigation, we feel that it is important to explicitly discuss in the paper a set of ethical considerations due to the potential sensitivity of facial information and any further investigation or practical use of this technology will be carefully tested to ensure that the training data is representative of the intended user population.”