Research on Narrowband Communications

Abstract

This document reports on work toward a very low rate vocoder. We model speech as a Markov Chain of spectral templates for the unsupervised learning approach to very low rate vocoding. This quarter we compared several clustering techniques. We determined that a hierarchical approach to clustering is economical with minimal loss in performance. Furthermore, we found that a small number of spectral templates (from 128 to 256) is sufficient for vocoding with good intelligibility. Also, in the phoneme recognition approach, we automated the training of the diphone network on labelled speech.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 1980
Accession Number
ADA102706

Entities

People

  • John Makhoul
  • Richard Schwartz
  • Salim Roukos

Organizations

  • BBN Technologies

Tags

Communities of Interest

  • Energy and Power Technologies
  • Human Systems

DTIC Thesaurus Topics

  • Algorithms
  • Classification
  • Clustering
  • Complex Systems
  • Computer Programming
  • Contracts
  • Databases
  • Debugging
  • Intelligibility
  • Language
  • Markov Chains
  • Markov Models
  • Recognition
  • Speech
  • Statistics
  • Trees (Data Structures)
  • Unsupervised Machine Learning

Fields of Study

  • Computer science

Readers

  • Mathematical Modeling and Probability Theory.
  • Speech Processing/Speech Recognition.

Technology Areas

  • AI & ML