scispace - formally typeset
Search or ask a question

Answers from top 9 papers

More filters
Papers (9)Insight
This paper proposes a novel approach to voice conversion with non-parallel training data.
Presents a new voice conversion algorithm.
In the experiments, we achieve high-quality converted voice, that performs equally well or better than mono-lingual voice conversion.
Proceedings ArticleDOI
Chen Zhi, Zhang Linghua 
01 Nov 2010
4 Citations
This capability is very fit of voice conversion.
The results show that humans are able to perceive empathy and emotions in robot speech, and prefer it over the standard robotic voice.
Book ChapterDOI
Robert Vích, Martin Vondra 
21 Feb 2011
6 Citations
The proposed voice conversion procedure results in speech with high naturalness.
By utilizing these approaches, the proposed method can change the spectrum and the prosody for an emotional voice at the same time, and was able to outperform other state-of-the-art methods for emotional voice conversion.
Experimental results showed that our method obtained higher similarity and naturalness than the best non-parallel voice conversion method in Voice Conversion Challenge 2018.
The results of the listening tests indicate that the proposed voice transformation provides better mapping of the voice characteristics compared to the earlier method proposed by the author.

Related Questions

How can tts be made less robotic?5 answersTo make Text-to-Speech (TTS) less robotic, various approaches have been proposed in recent research. One method is to enhance naturalness and efficiency by utilizing non-autoregressive models like Parallel Tacotron, which incorporates a variational autoencoder-based residual encoder for improved naturalness and parallelizability. Another strategy involves replacing the attention mechanism with an explicit duration predictor, as seen in the Non-Attentive Tacotron, which allows for better control over duration and enhances robustness in TTS systems. Additionally, leveraging syntactically parsed trees to organize inter-phrase/word information has shown to improve pronunciation clarity, prosody, and perceived naturalness in synthesized speech, as demonstrated in a study by Guo et al.. By combining innovative techniques like these, TTS systems can achieve higher quality, more human-like speech output.
What are the differences between whispered and normal voice?4 answersWhispered speech differs from normal speech in several ways. First, whispered speech lacks a periodic component called the Fundamental Frequency, which characterizes the pitch. Second, whispered speech has significantly lower energy compared to normal speech, making it more susceptible to external noise. Third, whispered speech can be used as a semi-silent speech interaction in public places without being audible to others. Additionally, whispered speech can be converted to normal speech using various techniques, such as generative adversarial network-based models, autoencoder-based frameworks, and bidirectional long short-term memory-based architectures. Converting whispered speech to normal speech can improve speech quality and intelligibility, making it beneficial for people with speech or hearing impairments.
How do I edit a robots txt file?8 answers
How do I transfer war robots from Android to PC?7 answers
How do you mod war robots?4 answers
How to convert normal amplifier to Bluetooth?9 answers