|
|
Lze použít automatické rozpoznávání řeči k hodnocení kvality řeči?
Nouza, Jan ; Vích, Robert ; Vondra, Martin
In the contribution several case studies are presented in which automatic speech recognition was tested as a means for evaluating of speech quality, either human or synthetic. Usually, speech quality is measured by subjective listening tests. Our aim is to investigate, whether these tests, which request considerable amount of human time and experience, could be replaced or supplemented by techniques based on ASR.
|
|
|
Konverze pohádkových hlasů pro TTS systém s kepstrálním popisem
Přibil, Jiří ; Přibilová, Anna
Our recent research in improvement of text-to-speech (TTS) synthesis was aimed at storytelling speaking style in addition to its multi-voice realization and expression of emotional states. Storytelling speaking style is suitable for applications aimed at children as well as applications aimed at blind people. In this contribution the experiments with the storytelling voice conversion performed on the short sentences of stories in Slovak and Czech are described.
|
|
|
Návrh vhodných prozodických modelů pro dialogové systémy
Horák, Petr
This paper deals with the improving of the synthetic prosody modeling especially with the improving of the intonation modeling. A mathematical model of the pitch contour modeling can significantly limit the complexity of intonation rules creation and increase the naturalness of resulting synthetic speech. The linear prediction intonation model implemented in TTS system Epos uses excitation by rules and provides in conjunction with a triphone time domain inventories more naturalness synthetic speech.
|
|
|
Současný stav vývoje českého TTS systému EPOS
Chaloupka, Zdeněk ; Horák, Petr
This contribution is focused on the current state of the Epos Text-To-Speech (TTS) system. Recently, a MBROLA-like synthesis interface has been developed. This interface synthesizes speech phone by phone, so the length and prosody points of each phone are specified. Several problems were encountered while performing MBROLA-like synthesis. These problems are associated with the speech inventory, which is primarily designed for the time domain Pitch-Synchronous OverLap-and-Add (PSOLA) synthesis.
|
|
| |
|
| |
|
|
Možnosti modelování prozodie TTS systému Epos s použitím MBROLA rozhraní
Horák, Petr ; Chaloupka, Zdeněk
This paper deals with prosody modelling possibilities of the Epos Text-To Speech (TTS) sytem. The aim is to build the TTS system with an absolute duration modelling of the separate sounds and the possibility of using MBROLA compatible Czech voices. The Epos is a very flexible language independent TTS system. It can be widely configured without the need of recompilation. The use of editable rules for every step during the synthesis makes it easy to monitor the progress of the synthesis and to apply changes.
|
|
|
Použití RLPC inventářů systému Festival v Eposu
Chaloupka, Zdeněk ; Horák, Petr
The aim of this paper is to describe a possibility of the new voices implementation into the Epos text-to-speech (TTS) system. We implemented voices from the Festival TTS system. This system synthesizes text from speech units, which are stored in an inventory file as Residual Linear Prediction Coding (RLPC) coefficients. The inventory file provides every information needed for the text synthesis. The text is synthesized in the MROLA format, thus a phoneme length (and a prosody) can be determined directly.
|
|
|
Hodnocení změny hlasu v TTS systému s kepstrálním popisem
Přibil, Jiří ; Přibilová, Anna
Voice conversion, i.e. modification of a speech signal to sound as if spoken by a different speaker, finds its use in speech synthesis with a new voice without necessity of a new database. It is very useful in special aids for blind and partially sighted people, e.g. Braille notetaker based on the Pocket PC. This paper evaluates voice conversion implemented in the Czech and Slovak text-to-speech (TTS) system based on cepstral description of speech database using the source-filter speech model.
|
|
|
Frequency scale mapping methods for voice conversion
Přibilová, Anna
The research report compares several approcaches to frequency scale mapping for voice conversion. All the presented methods may be used for transformation of voice characteristics between male and female or childisch. The advantage of the proposed non-linear methods is possibility of their implementation in the text-to-speech systems with cepstral description because of the unchanged number of points of the modified speech spectrum envelope in FFT-based methods.
|
guest ::
