Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Efficient high-order hidden Markov modelling
Dissertation (Ph. D.) -- University of Stellenbosch, 1997.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
Stellenbosch : Stellenbosch University
2012
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867614012676505600 |
|---|---|
| access_status_str | Open Access |
| author | Du Preez, Johan Adam |
| author2 | Barnard, E. |
| author_browse | Barnard, E. Du Preez, Johan Adam |
| author_facet | Barnard, E. Du Preez, Johan Adam |
| author_sort | Du Preez, Johan Adam |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Dissertation (Ph. D.) -- University of Stellenbosch, 1997. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/55523 |
| institution | Stellenbosch University (South Africa) |
| language | English |
| last_indexed | 2026-06-10T12:45:16.097Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2012 |
| publishDateRange | 2012 |
| publishDateSort | 2012 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/55523 Efficient high-order hidden Markov modelling Du Preez, Johan Adam Barnard, E. Weber, D. M. Stellenbosch University. Faculty of Engineering. Dept. of Electrical & Electronic Engineering. Automatic speech recognition Markov processes Dissertations -- Electronic engineering Dissertation (Ph. D.) -- University of Stellenbosch, 1997. Currently, first-order hidden Markov models (HMMs) form the backbone around which most automatic speech processing applications are built. Their higher-order extensions are known to be more powerful, but, due to their complexity and computational demands, they are seldomly used. It is the purpose of this work to advance their application. In this work we unify HMMs of all orders by deriving and proving the Order rEDucing (ORED) algorithm. This algorithm, will reduce any higher-order HMM (also mixed-order) to an equivalent first-order representation. This makes it possible to process any higher-order HMM using known first-order algorithms, thereby making unnecessary the current approach to extending specific HMM algorithms to specific higher orders. It also provides an alternative theoretical basis to reason about high-order HMMs. From this perspective high-order transition probabilities are simply powerful mathematical specifications of first-order topology. We use this insight to explain old topologies and to design new ones. We address computational concerns by developing the Fast Incremental Training (FIT) algorithm. This algorithm avoids training redundant high-order probabilities by noting which lower-order transition probabilities are zero. This considerably reduces the memory and processor requirements during training. In addition, the resultant models have far fewer parameters and generalise better on previously unseen data. To share the practical applicability of our methodology we apply it to automatic language recognition. We find that it compares well with systems that require expensive transcribed databases (ours system does not require this). Doctoral 2012-08-27T11:37:06Z 2012-08-27T11:37:06Z 1997 Thesis http://hdl.handle.net/10019.1/55523 en Stellenbosch University 166 pages : ill. application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Automatic speech recognition Markov processes Dissertations -- Electronic engineering Du Preez, Johan Adam Efficient high-order hidden Markov modelling |
| title | Efficient high-order hidden Markov modelling |
| title_full | Efficient high-order hidden Markov modelling |
| title_fullStr | Efficient high-order hidden Markov modelling |
| title_full_unstemmed | Efficient high-order hidden Markov modelling |
| title_short | Efficient high-order hidden Markov modelling |
| title_sort | efficient high order hidden markov modelling |
| topic | Automatic speech recognition Markov processes Dissertations -- Electronic engineering |
| url | http://hdl.handle.net/10019.1/55523 |
| work_keys_str_mv | AT dupreezjohanadam efficienthighorderhiddenmarkovmodelling |