Home > Search Results: Nekvinda, Tomáš
Search:
Search Tips :: Advanced Search
Search collections:
Sort by: Display results: Output format:
National Repository of Grey Literature 2 records found  Search took 0.00 seconds. 
Multilingual speech synthesis
 Nekvinda, Tomáš ; Dušek, Ondřej (advisor) ; Peterek, Nino (referee)
This work explores multilingual speech synthesis. We compare three models based on Tacotron that utilize various levels of parameter sharing. Two of them follow recent multilingual text-to-speech systems. The first one makes use of a fully-shared encoder and an adversarial classifier that removes speaker-dependent information from the encoder. The other uses language-specific encoders. We introduce a new approach that combines the best of both previous methods. It enables effective parameter sharing using a meta- learning technique, preserves encoder's flexibility, and actively removes speaker-specific information in the encoder. We compare the three models on two tasks. The first one aims at joint multilingual training on ten languages and reveals their knowledge-sharing abilities. The second concerns code-switching. We show that our model effectively shares information across languages, and according to a subjective evaluation test, it produces more natural and accurate code-switching speech.
Detailed record
Fixing of Facial Triangle Meshes
 Nekvinda, Tomáš ; Pelikán, Josef (advisor) ; Krajíček, Václav (referee)
The purpose of this work is to create an algorithm for automatic cleaning and trimming of three-dimensional facial scans. In view of that, we have developed an algorithm which consists of three central parts. The frst part is a novel landmark detection algorithm based on discrete diferential geometry and machine learning methods. The second part is represented by a method removing defects, spikes, and blobs and the last part is an enhancement of an algorithm for hole flling. The outcome of this work is a program which can automatically clean and trim three-dimensional facial scans and moreover, it can detect nose tip, nose root, and mouth and eye corners. As our testing has shown, the program performs well on facial scans produced by the optical scanner Vectra3D.
Detailed record

Interested in being notified about new results for this query?
Subscribe to the RSS feed.
Digital Repository :: Search :: Submit :: Personalize :: Help
Powered by Invenio v1.1.2
Maintained by admin.invenio@techlib.cz
This site is also available in the following languages:
Česky  English