Pune, India

Shrikant Joshi


Average Co-Inventor Count = 5.2

ph-index = 1

Forward Citations = 1(Granted Patents)


Location History:

  • Pune, IN (2019)
  • Kolhapur, IN (2023)

Company Filing History:


Years Active: 2019-2025

Loading Chart...
4 patents (USPTO):Explore Patents

Title: Shrikant Joshi: Innovating Multilingual Transcription with Cutting-Edge System and Method

Introduction:

In the ever-evolving world of language and communication, Shrikant Joshi, a Pune-based inventor, has made remarkable strides in revolutionizing the transcription of spoken words. With his groundbreaking patent, titled "System and Method for Transcription of Spoken Words using Multilingual Mismatched Crowd Unfamiliar with a Spoken Language," Joshi has paved the way for efficient multi-scripted transcription processes. Let us delve into the details of his remarkable work and explore the company he works for, Tata Consultancy Services Limited.

The Patent Innovation:

Joshi's inventive system and method focus on transcription techniques that employ a diverse pool of multilingual workers who are unfamiliar with the spoken language being transcribed. By leveraging this multilingual mismatched crowd, Joshi's patent addresses the challenges of transcription in various languages, resulting in a more adaptable and accurate system.

Key Features:

The patent outlines an intricate process that encompasses several crucial components. Firstly, the system collects multi-scripted noisy transcriptions of spoken words from workers within the multilingual mismatched crowd. These workers are selected for their unfamiliarity with the source language, ensuring unbiased transcription results.

Next, the collected words are mapped to a phoneme sequence in the source language utilizing script-specific graphemes to phoneme models. This step allows for accurate representation of spoken words across different languages and scripts.

Joshi's system also incorporates the creation of a multi-scripted transcription script, which incorporates worker-specific and global insertion-deletion-substitution (IDS) channels. These features optimize the transcription process by accounting for phoneme variations and ensuring the utmost accuracy and fidelity in the final transcriptions.

Worker Reputation and Task Allocation:

To further enhance the transcription process, the patent introduces a reputation-based worker allocation system. This system determines the reputation of workers based on word belief, a metric derived from the likelihood probability of mapped phoneme sequences of transcriptions. The allocation of transcription tasks is then based on these reputation scores, ensuring that the most reliable workers handle critical transcription assignments.

Shrikant Joshi at Tata Consultancy Services Limited:

Shrikant Joshi's impressive expertise aligns with his position at Tata Consultancy Services Limited, a renowned global IT services and consulting company. With a strong focus on innovation and cutting-edge solutions, TCS provides a conducive environment for inventors like Joshi to thrive and bring their inventive ideas to life.

Collaborators and Coworkers:

Joshi has had the privilege of working alongside talented individuals who have contributed to his groundbreaking work. Some of his noteworthy coworkers include Purushotam Gopaldas Radadia and Kanika Kalra. Together, they have been instrumental in the implementation and development of the transcription system, making remarkable strides in the field of multilingual transcription.

Conclusion:

Shrikant Joshi's patent on the system and method for multilingual transcription is a testament to his ingenuity and passion for advancing language processing technologies. His innovative approach to leveraging multilingual, mismatched crowds has the potential to transform the transcription landscape. Through his association with Tata Consultancy Services Limited and collaboration with talented professionals, Joshi is paving the way for more efficient and accurate transcription processes across languages and scripts.

This text is generated by artificial intelligence and may not be accurate.
Please report any incorrect information to support@idiyas.com
Loading…