Intermediate-state HMMs to capture continuously-changing signal features

Gustav Eje Henter, W. Bastiaan Kleijn

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

Traditional discrete-state HMMs are not well suited for describing steadily evolving, path-following natural processes like motion capture data or speech. HMMs cannot represent incremental progress between behaviors, and sequences sampled from the models have unnatural segment durations, unsmooth transitions, and excessive rapid variation. We propose to address these problems by permitting the state variable to occupy positions between the discrete states, and present a concrete left-right model incorporating this idea. We call this intermediate-state HMMs. The state evolution remains Markovian. We describe training using the generalized EM-algorithm and present associated update formulas. An experiment shows that the intermediate-state model is capable of gradual transitions, with more natural durations and less noise in sampled sequences compared to a conventional HMM.
Original languageEnglish
Title of host publicationProc. Interspeech 2011
Place of PublicationFlorence, Italy
Number of pages4
Publication statusPublished - 1 Aug 2011
EventInterspeech 2011- 12th annual Conference of the International Speech Communication Association - Florence, Italy
Duration: 27 Aug 201131 Aug 2011


ConferenceInterspeech 2011- 12th annual Conference of the International Speech Communication Association

Keywords / Materials (for Non-textual outputs)

  • Markov models
  • HMMs
  • speech synthesis


Dive into the research topics of 'Intermediate-state HMMs to capture continuously-changing signal features'. Together they form a unique fingerprint.

Cite this