White Rose University Consortium logo
University of Leeds logo University of Sheffield logo York University logo

Personalising synthetic voices for individuals with severe speech impairment.

Creer, Sarah M (2010) Personalising synthetic voices for individuals with severe speech impairment. PhD thesis, University of Sheffield.

[img]
Preview
Text (Thesis)
522463.pdf
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (8Mb) | Preview
[img] Audio
3-1_lumberjack.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (93Kb)
[img] Audio
3-2_track86f.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (62Kb)
[img] Audio
3-3_part33_klatt.au
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (266Kb)
[img] Audio
3-4_part35_dectalk.au
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (288Kb)
[img] Audio
3-5_time-10_35am.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (42Kb)
[img] Audio
3-6_intro_fest.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (93Kb)
[img] Audio
3-7_original_smc.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (42Kb)
[img] Audio
3-8_fest_cupoftea.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (15Kb)
[img] Audio
3-9_mt_cupoftea.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (19Kb)
[img] Audio
3-10_sp1_arctic_a0586.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (41Kb)
[img] Audio
3-11_bad_fest.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (35Kb)
[img] Audio
3-12_sp1_fest.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (17Kb)
[img] Audio
3-13_hts_cupoftea.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (27Kb)
[img] Audio
4-1_sp5_a0251.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (42Kb)
[img] Audio
5-1_sp1_arctic_a0183.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (42Kb)
[img] Audio
5-2_sp2_arctic_a0290.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (29Kb)
[img] Audio
5-3_ave_test_arctic_a0183.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (20Kb)
[img] Audio
5-4_sp1_10_test_arctic_a0183.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (21Kb)
[img] Audio
5-5_sp1_100_test_arctic_a0183.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (20Kb)
[img] Audio
5-6_sp1_500_test_arctic_a0183.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (20Kb)
[img] Audio
5-7_sp1_target_test_arctic_a0183.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (22Kb)
[img] Audio
5-8_ave_test_arctic_a0290.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (21Kb)
[img] Audio
5-9_sp2_10_test_arctic_a0290.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (27Kb)
[img] Audio
5-10_sp2_100_test_arctic_a0290.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (27Kb)
[img] Audio
5-11_sp2_500_test_arctic_a0290.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (29Kb)
[img] Audio
5-12_sp2_target_test_arctic_a0290.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (26Kb)
[img] Audio
6-1_sp3_arctic_a0064.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (65Kb)
[img] Audio
6-2_sp4_arctic_a0050.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (33Kb)
[img] Audio
6-3_sp5_arctic_a0359.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (88Kb)
[img] Audio
6-4_sp5_allfeats.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (85Kb)
[img] Audio
6-5_sp5_edited_allfeats.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (62Kb)
[img] Audio
6-6_peter_acapela_short.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (50Kb)
[img] Audio
6-7_sp3_3_scribe_sent013.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (54Kb)
[img] Audio
6-8_sp4_3_scribe_sent019.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (56Kb)
[img] Audio
6-9_sp5_3_scribe_sent088.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (56Kb)
[img] Audio
6-10_sp5_9_scribe_sent088.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (69Kb)
[img] Audio
6-11_sp3_5_scribe_sent013.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (69Kb)
[img] Audio
6-12_sp4_5_scribe_sent019.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (72Kb)
[img] Audio
6-13_sp3_6A_scribe_sent013.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (54Kb)
[img] Audio
6-14_sp4_6A_scribe_sent019.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (56Kb)
[img] Audio
6-15_sp5_6A_scribe_sent088.mp3
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (56Kb)
[img] Text
thesis_sound_files.html
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (14Kb)

Abstract

Speech technology can help individuals with speech disorders to interact more easily. Many individuals with severe speech impairment, due to conditions such as Parkinson's disease or motor neurone disease, use voice output communication aids (VOCAs), which have synthesised or pre-recorded voice output. This voice output effectively becomes the voice of the individual and should therefore represent the user accurately. Currently available personalisation of speech synthesis techniques require a large amount of data input, which is difficult to produce for individuals with severe speech impairment. These techniques also do not provide a solution for those individuals whose voices have begun to show the effects of dysarthria. The thesis shows that Hidden Markov Model (HMM)-based speech synthesis is a promising approach for 'voice banking' for individuals before their condition causes deterioration of the speech and once deterioration has begun. Data input requirements for building personalised voices with this technique using human listener judgement evaluation is investigated. It shows that 100 sentences is the minimum required to build a significantly different voice from an average voice model and show some resemblance to the target speaker. This amount depends on the speaker and the average model used. A neural network analysis trained on extracted acoustic features revealed that spectral features had the most influence for predicting human listener judgements of similarity of synthesised speech to a target speaker. Accuracy of prediction significantly improves if other acoustic features are introduced and combined non-linearly. These results were used to inform the reconstruction of personalised synthetic voices for speakers whose voices had begun to show the effects of their conditions. Using HMM-based synthesis, personalised synthetic voices were built using dysarthric speech showing similarity to target speakers without recreating the impairment in the synthesised speech output.

Item Type: Thesis (PhD)
Academic Units: The University of Sheffield > Faculty of Engineering (Sheffield) > Computer Science (Sheffield)
The University of Sheffield > Faculty of Science (Sheffield) > Computer Science (Sheffield)
Identification Number/EthosID: uk.bl.ethos.522463
Depositing User: EThOS Import Sheffield
Date Deposited: 11 Sep 2019 11:59
Last Modified: 11 Sep 2019 11:59
URI: http://etheses.whiterose.ac.uk/id/eprint/21830

You do not need to contact us to get a copy of this thesis. Please use the 'Download' link(s) above to get a copy.
You can contact us about this thesis. If you need to make a general enquiry, please see the Contact us page.

Actions (repository staff only: login required)