Trint App Offers Outstanding Accuracy In Speech To Text Transcription

News about breakthrough artificial intelligence systems often creates a worry in the minds of people. It deals with the fear of robots taking up jobs. However, again, at times the work becomes so monotonous that people appear happy if robots performed those task. There are people who simply dislike the concept of transcribing audio. Thankfully, there is an app to deal with this known as Trent app.

The Trent is a superb text to speech web app that listens to long blocks of audio and converts it into texts. In fact, it does the transcription rather flawlessly. To be precise, it is a portmanteau of interview and transcription. Apart from these, the app can also neatly distinguish between several people in recordings, lets users assign time code to the transcription for reference at a later time, Digital Trends reported.

"We use the best automated speech-to-text you will find," CEO and co-founder Jeffrey Kofman. He further added, "With reasonably clear audio we can return transcripts that are 95-98 percent accurate. When you take very clear speakers like Trump or Obama, our automated transcripts are often 99 percent accurate. People tell us they think what we've built is magic."

As per Quartz, a former journalist, Kofman also claimed that it is important for lawyers, researchers and others to know if they can trust the transcripts. Due to this, the Trint app combines two softwares. It includes tools that not only performs automated speech to text, but also provides an easy way to verify and search as well as correct the output.

The software includes both an audio/video player and a text editor. Interestingly, it allows the users to check the finished and final product as a karaoke track with both audio and video on screen. This helps users spot the error easily if there are any. The Trint app also transcribes in several languages including Australian English, North American and British English. The product is available for $10 to $15 an hour depending on the audio type.