Tejas Godambe

CV | Github | Google-Scholar | LinkedIn | Contact

I work as a Speech Research Engineer at Genesys, Hyderabad, India where my primary area of R&D is Neural Network based Automatic Speech Recognition. Prior to joining Genesys, I obtained MS by Research from International Institute of Information Technology, Hyderabad (IIIT-H) under the supervision of Dr. Kishore Prahallad and Dr. Suryakanth Gangashetty. The topic of my Masters thesis was Speech recognition based confidence measures for building voices from untranscribed speech. Before joining IIIT-H, I worked as a Project Assistant with Dr. Samudravijaya K on a DIT sponsored project "Speech-based access for agricultural commodity prices in Marathi language" at Tata Institute of Fundamental Research, Mumbai, India.

Published Papers

  1. ININ submission to Zero Cost ASR at MediaEval 2016 (pdf)
    Tejas Godambe, Naresh Kumar, Pavan Kumar, Veera Raghavendra, Aravind Ganapathiraju
    The 28th ACM Conference on Hypertext and Social Media (HT), Jul 4-7, Prague, Czech Republic, 2017
  2. Developing a unit selection voice given audio without corresponding text (arxiv, pdf, slides, poster, bibtex, code)
    Tejas Godambe, Sai Krishna Rallabandi, Suryakanth V Gangashetty, Ashraf Alkhairy and Afshan Jafri
    The 39th European Conf. on Information Retrieval (ECIR), Apr 9-13, Aberdeen, Scotland UK, 2017
  3. Data pruning and objective as-sessment of intelligibility using con dence measures for unit selection synthesis system (arxiv, pdf, slides, poster, bibtex, code)
    Tejas Godambe, Sai Krishna Rallabandi, Suryakanth V Gangashetty
    NIPS Workshop on Interpretable Machine Learning in Complex Systems, Dec 9, Barcelona, Spain, 2016
  4. Adaptation of acoustic models for im-proved Marathi speech recognition (pdf, slides, poster, bibtex, code)
    Tejas Godambe, Namrata Karkera, Samudravijaya K
    The 39th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR), Jul 17-21, Pisa, Italy, 2016
  5. Speech data acquisition for voice based agricultural in-formation retrieval (pdf, slides, poster, bibtex, code)
    Tejas Godambe, Samudravijaya K
    The 25th International World Wide Web Conference (WWW), Apr 11-25, Montreal, Canada, 2016

Preprints

  1. Improving Distributed Representations of Tweets - Present and Future (2017)
    Ganesh J
  2. Interpretation of Semantic Tweet Representations (2017) (arxiv, pdf, bibtex, code)
    Ganesh J, Manish Gupta, Vasudeva Varma
  3. Discovery and Interpretation of Embedding Models for Knowledge Representation (Master's Thesis) (2016) (pdf, slides)
    Ganesh J

Softwares

  1. Singular Value Decomposition based DNN compression (code, reference)
  2. Distlling the Knowledge in a Neural Network (code, reference)
  3. Ensemble of Neural Networks (code, reference)
  4. Conv Nets for Speech Recognition (code)
  5. Concatenation of phone/word instances for auditory verification of the goodness of alignments (code)
Back to top