In this paper we introduce a set of resources and tools aimed at providing support for natural language processing, text-to-speech synthesis and speech recognition for Romanian. While the tools are general purpose and can be used for any language (we successfully trained our system for more than 50 languages and participated in the Universal Dependencies Shared Task), the resources are only rel...

Current speech synthesis methods typically operate on isolated sentences and lack convincing prosody when generating longer segments of speech. Similarly, prevailing TTS evaluation paradigms, such as intelligibility (transcription word error rate) or MOS, only score sentences in isolation, even though overall comprehension arguably is more important for speech-based communication. In an effort ...

Speech contains prosodic cues such as pauses between different phrases of a sentence. These intonational phrase boundaries (IPBs) elicit a specific component in event-related brain potential studies, the so-called closure positive shift. The aim of the present functional magnetic resonance imaging study is to identify the neural correlates of this prosody-related component in sentences containi...

One approach to the generation of natural-sounding synthesized speech waveforms is to select and concatenate units from a large speech database. Units (in the current work, phonemes) are selected to produce a natural realisation of a target phoneme sequence predicted from text which is annotated with prosodic and phonetic context information. We propose that the units in a synthesis database ca...

This study tested the importance of F0 variation for tone language comprehension. The intelligibility of Mandarin sentences with natural F0 contours was compared to the intelligibility of monotone (flat-F0) sentences created via speech resynthesis. In a quiet background, flat-F0 speech was just as intelligible as natural speech (about 94% intelligible), highlighting the robustness of the langua...