Harnessing the strength of Machine Learning (ML), this newly launched app provides fine speech-to-text services to large group of people
Hearing is one of the vital senses for a human being that enables us to take notice of a certain mesage, understand it and convey back one’s own thoughts and opinions to the surrounding world.
However, due to factors such as ageing, ear infections, recreational activities or accidents, an individual’s ability to hear gets adversely affected.
This causes an inability to effectively communicate with others. Several hearing aid devices do manage to provide some assistance to those with mild or severe hearing impairment. Yet, they cannot be suitably used for each and every case and not all of them are easy to use.
Google’s unique solution
The tech giant earlier this year launched an app called the Live Transcribe that facilitates people with hearing problems to connect with the world around them by transcribing conversations in real-time. The app uses a host of tools, both hardware and software, to enable people to follow and engage in a conversation that they otherwise would have been excluded from.
Sagar Savla, Google product manager claimed that the app works in real time and helps in interpreting the spoken words.
Besides the above use case, the Live Transcribe app can also be used by students and journalists to take notes.
Powered by Machine Learning
Google’s Live Transcribe app is pretty simple to use. A phone’s microphone listens to a particular audio (human speech or any other sound), which the app analyses with the help of Machine Learning and then transcribe that sound into text. The user then can read the relevant text and get the message.
The app’s Machine learning (ML) model comprises of 3 parts:-
- The Acoustic model converts the sounds to phonemes (units of sound that distinguish one word from another in a particular language).
- The Pronunciation model converts those phonemes into words.
- The Language model then adds context to the words to make sense of the entire conversation.
Savla adds that the app adeptly understands the context of a converstaion. So if you are talking about buying a new jersey in New York, it actually understands the capitalisation and makes the jersey smaller so that it knows that you are talking about the shirt and not the place.
Large-scale implementation
Live Transcribe, which is available for Android devices, is accessible in 150 countries across the world and can transcribe up to 70 different languages including Hindi, Portuguese, Spanish and Italian…all in real time. In addition to that, the Live Transcribe app can also recognise 40 different sound effects like car sirens/horns, clapping, music, applause, shouting, speeding vehicle and baby crying among others.
Users can also copy and save transcripts of conversation on their devices for three days.
