“Application to directly convert songs aaAainto Guitar / Piano tabs“

Gurgaon, India.Jan 22,2019 - In a major development, ​Acousterr​, an innovative online platform for music learners, has launched an artificial intelligence application for the automatic transcription of audio files to guitar / piano tablature.

Acousterr, an Indian startup, based in Gurgaon which saw the light of day in Nov 2017 provides extremely powerful ​online tab maker and editor tool to create and playback tablature. Currently thousands of music learners and teachers are creating tabs to help acquire proficiency in playing guitar/piano/ukulele. Acousterr takes pride in having a good collection of guitar applications that are helpful to both learners and professionals.

The latest AI application to automatically ​transcribe music audios to guitar /piano tabs would benefit music enthusiasts, both beginners who depend on tabs to learn a song as well as advanced learners who spend hours to transcribe a song by ear.

Giving details of the product, the founders stated that it is based on modern day Deep Learning technology which allows computational models that are composed of multiple processing layers to learn representations of data with multiple levels of abstraction. A state-of-the-art Deep Learning application is focused on training deep (many layered) neural network models using the back propagation algorithms. Deep Learning employs structures in layers (artificial neural network) to create an artificial intelligence that can learn and make intelligent decisions on its own. As such, the product developed by Acousterr to automatically transcribe audio files to tabs provides a reasonable accuracy to get started. During the process of transcription, the output of AI can be visualised on a post processing editor, which is like an enhanced spectrogram analyzer, and can be edited effectively to correct errors easily and make sense of the results.

The beta version of the product was launched in 2018 and has drawn overwhelming response from music enthusiasts. Currently tha AI works well for


music pieces having a single predominant instrument. In a polyphonic music piece, the results expected are highly subjective, which makes it an extremely difficult problem to solve. However with the incorporation of post processing spectrogram editor, sufficient power is given to the user to interpret and correct the result. Overall it is a very powerful system and aims to make music learning easier, and will continue to get better with more data at hand.

The founders Nischey Grover (Alumnus IIT Roorkee) and Rashmi Chaurasiya (Doctoral Researcher) who themselves are Deep Learning enthusiasts, expressed their optimism about the adoption of technology and believe that such technologies can make a music learner’s life simpler. The application is hosted at their website acousterr.com and is free for use for a limited time. They can be contacted below, and they are willing to receive any enquiries through their email contact included below.

email : ​[email protected]

website : ​https://www.acousterr.com

phone : +91 8884387325