Paper explaining training DNNs using Probabilistic Transcripts
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
poster
README.md
template.pdf

README.md

Title: An investigation on training deep neural networks using probabilistic transcriptions

Authors: Amit Das, Mark Hasegawa-Johnson

Status: Published in Interspeech 2016

Abstract:

In this study, a transfer learning technique is presented for cross-lingual speech recognition in an adverse scenario where there are no natively transcribed transcriptions in the target language. The transcriptions that are available during training are transcribed by crowd workers who neither speak nor have any familiarity with the target language. Hence, such transcriptions are likely to be inaccurate. Training a deep neural network (DNN) in such a scenario is challenging; previously reported results have described DNN error rates exceeding the error rate of an adapted Gaussian Mixture Model (GMM). This paper investigates multi-task learning techniques using deep neural networks which are suitable for this scenario. We report, for the first time, absolute improvement in phone error rates (PER) in the range 1.3-6.2% over GMMs adapted to probabilistic transcriptions. Results are reported for Swahili, Hungarian, and Mandarin.

Code: