Testing voice technology from the seabed of the Titanic

During an expedition to the sunken Titanic, computer scientist Alex Waibel tested voice technology with video capability from a submarine. It sent texts to the surface via sonar, which were then converted into spoken language and video using artificial intelligence (AI). The Karlsruhe Institute of Technology (KIT) researcher told the German Press Agency that they had passed some of the dialogues. “We could see that it really works.”

The Titanic sank on her maiden voyage in 1912. The passenger ship lies at a depth of around 3800 meters in the Atlantic. Sparks aren’t possible from there because of the salt water masses, Waibel explained. Sonar, however, only allows for “incredibly low bandwidth”. That’s just enough for texting.

The technology being tested works as follows: Before the dive, Waibel and his colleagues recorded videos and voice samples of themselves. If text messages now reach the computer system, the AI ​​implements this in such a way that it looks and sounds on the video as if the person himself is speaking – including lip movements.

What looks like a PR gimmick by tech-savvy scientists, particularly on the Titanic expedition, has a serious background: “There are enough places in the world where the tape bandwidth is so poor that only text transmission is possible,” Waibel said. With new technology, video communication may one day become possible.

But the mission also revealed the pitfalls: One of the two sonar devices failed, Waibel said. Therefore, only part of the dialogue could still be transmitted from the submarine. He also came up with new ideas: U-boat crews worked a lot with abbreviations to compress texts. Another goal is to reduce the technology so that it fits in a pocket. Overall, Waibel was satisfied: “A good start has been made.”

Incidentally, one of the biggest challenges when converting texts into videos has nothing to do with language, the scientist reveals: “If the person doesn’t say anything, it’s surprisingly difficult.” Then the lips in the videos hardly move at all. .

Waibel was part of a larger mission involving biologists and archaeologists, among others. There are always such expeditions on the Titanic.

For more than 30 years, the researcher has been interested in AI and machine learning in language and communication technologies. Among other things, he developed the world’s first automatic simultaneous translation service at a university, according to KIT. The “Lecture Translator” automatically records the speaker’s lecture and simultaneously translates the voice cues into English, which are then displayed as subtitles. Students without knowledge of German can take the course via laptop, smartphone or tablet.

Mathew Baynton

"Bacon nerd. Extreme zombie scholar. Hipster-friendly alcohol fanatic. Subtly charming problem solver. Introvert."

Leave a Reply

Your email address will not be published.