Esther Klabbers and Raymond Veldhuis, "On the reduction of concatenation artefacts in diphone synthesis, " in Proceedings of ICSLP, December 1998. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Corpus-Based Unit Selection for Natural-Sounding Speech Synthesis - Yi (2003) (Correct)
....Much of the earlier work in the literature has concentrated on instance level costs that directly compare speech segments, or instantiations of the speech units. Numerical metrics such as Euclidean, Kullback Leibler, and Mahalanobis distances calculated over spectral features have been considered [21, 22, 39, 61, 69, 160, 137, 70, 161]. The concatenation cost defined here bears similarity to disconcatibility as proposed by Iwahashi [63] and to splicing cost proposed as by Bulyko [22] The use of mutual information to find boundaries across which information is blocked is related to rifts [6] as studied in statistical machine ....
.... predict which human voices are more suitable for concatenative speech synthesis [142, 141] Psychoacoustic experiments at a 144 smaller time scale have investigated whether discontinuities at concatenation boundaries are perceptible and what measures can be used to predict signal discontinuities [39, 69, 160, 137]. 6.2 Processes Crucial to designing and building systems is an understanding of the underlying processes. This section is devoted to delineating the steps taken to rapidly prototype a system from scratch, including how to train a synthesizer for a new language and how to determine the set of ....
E. Klabbers and R. Veldhuis, "On the reduction of concatenation artefacts in diphone synthesis," in Proc. ICSLP '98, Sydney, Australia, Nov. 1998.
A New Distance Measure for Costing Spectral Discontinuities in.. - Donovan (2001) (5 citations) (Correct)
....determine how smoothly two segments will concatenate during synthesis. Various measures were used in the systems mentioned above, often based on distances between cepstra or mel cepstra, but in most cases little work appears to have been done to determine the usefulness of the measure selected. In [8] however, Klabbers Veldhuis evaluated a number of likely measures by comparing the numerical distances computed with them to listener ratings of discontinuities in a large number of concatenated stimuli. The measures included the Euclidean distance between ## and ## (formant frequency) pairs, the ....
....cannot be used because with 20 measures being evaluated there would be a high probability of one being judged significant just by chance. 5. Discussion As can be seen from Figure 1, the author s ratings correlate significantly with the Kullback Leibler distance (consistent with the result in [8]) the Mahalanobis distance between perceptual cepstra, the Mahalanobis distance between perceptual cepstra with deltas and delta deltas, and most significantly of all with the new distance measure described in Section 2. The high correlation of the author s ratings with Mahalanobis perceptual ....
Klabbers, E., and Veldhuis, R. (1998) On the Reduction of Concatenation Artefacts in Diphone Synthesis, Proc. ICSLP'98, Sydney.
Control of Spectral Dynamics in Concatenative Speech Synthesis - Wouters, Macon (2001) (13 citations) (Correct)
....methods can be successful only if acceptable concatenation points exist between the segments in the database. Furthermore, the selection may be suboptimal since the acoustic distance measures that are commonly used have only moderate correlation with human judgements of acoustic distortions [7] [8]. Other techniques have been proposed to mitigate the effects of concatenation artifacts by modifying the spectral characteristics of speech. Most approaches are based either on waveform interpolation of pitch periods or on smoothing of LPC derived parameters. In either approach, the region of ....
E. Klabbers and R. Veldhuis, \On the reduction of concatenation artefacts in diphone synthesis," in ICSLP, November 1998, vol. 6, pp. 2759-2762.
A Generic Algorithm for Generating Spoken Monologues - Klabbers, Krahmer, Theune (1998) Self-citation (Klabbers) (Correct)
No context found.
Klabbers, E. and Veldhuis, R., "On the Reduction of Concatenation Artefacts in Diphone Synthesis," 1998, these proceedings.
From Data to Speech: A General Approach - Theune, Klabbers, al. (2000) (10 citations) Self-citation (Klabbers) (Correct)
.... to express collocations (groups of words with a frozen meaning) Examples of collocations occurring in the GoalGetter templates are een doelpunt laten aantekenen From Data to Speech: A General Approach 15 ( have a goal noted ) as in Template Sent16) or de leiding nemen ( take the lead ) See Klabbers et al. 1998) for some further discussion. The syntactic trees in the templates are given in full detail because during prosody computation (see Section 3.2) they need to be converted into full metrical trees. The second element of a syntactic template is E: the slot fillers. Each open slot in the tree S is ....
....is a lot of variability in the realisation of the prosody. In order to achieve more natural sounding speech output, we are currently concentrating on the improvement of a few specific aspects of the diphone synthesis system, such as the occurrence of audible discontinuities at diphone boundaries (Klabbers Veldhuis, 1998) and duration control (Klabbers, 2000) 3.3.2 Phrase concatenation using prosodic variants Phrase concatenation is better suited for use in commercial systems because the speech quality is close to that of natural speech. Therefore, we chose this technique as the primary technique for speech ....
Klabbers, E., & Veldhuis, R. 1998. On the reduction of concatenation artefacts in diphone synthesis. In: Proceedings of ICSLP 1998.
A Generic Algorithm for Generating Spoken Monologues - Klabbers, Krahmer, Theune (1998) Self-citation (Klabbers) (Correct)
No context found.
Klabbers, E. and Veldhuis, R., "On the Reduction of Concatenation Artefacts in Diphone Synthesis," 1998, these proceedings.
Towards Phone Segmentation For Concatenative Speech Synthesis - Antonio (Correct)
No context found.
Esther Klabbers and Raymond Veldhuis, "On the reduction of concatenation artefacts in diphone synthesis, " in Proceedings of ICSLP, December 1998.
Unit Selection and Emotional Speech - Alan Black Language (2003) (2 citations) (Correct)
No context found.
E. Klabbers and R. Veldhuis, "On the reduction of concatenation artefacts in diphone synthesis," in ICSLP98, Sydney, Australia., 1998.
Perceptual And Objective Detection Of Discontinuities In - Concatenative Speech.. (Correct)
No context found.
E. Klabbers and R. Veldhuis, "On the reduction of concatenation artefacts in diphone synthesis," International Conference on Spoken Language Processing ICSLP 98, pp. 1983.
Data-Driven Perceptually Based Join Costs - Syrdal, Conkie (Correct)
No context found.
E. Klabbers and R. Veldhuis, "On the reduction of concatenation artefacts in diphone synthesis," International Conference on Spoken Language Processing ICSLP 98, pp. 1983.
Information-Theoretic Criteria for Unit Selection Synthesis - Yi, Glass (2002) (6 citations) (Correct)
No context found.
E. Klabbers and R. Veldhuis, "On the reduction of concatenation artefacts in diphone synthesis," Proc. ICSLP, 1983.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC