nascent
发表于 2025-3-23 13:04:54
http://reply.papertrans.cn/88/8765/876449/876449_11.png
史前
发表于 2025-3-23 17:04:39
Evaluation of the Impact of Corpus Phonetic Alignment on the HMM-Based Speech Synthesis Quality,equires phonetic symbols aligned to the speech corpus in order to train the models used for synthesis. Phonetic annotation is a complex task, since pronunciation usually differs from spelling, as well as differing among regional accents. In this paper, the infrastructure of a French TTS system is pr
愤慨一下
发表于 2025-3-23 20:46:31
http://reply.papertrans.cn/88/8765/876449/876449_13.png
abduction
发表于 2025-3-24 00:09:30
Combining Continuous Word Representation and Prosodic Features for ASR Error Prediction,r prediction in Automatic Speech Recognition (ASR) outputs and proposes to investigate the use of continuous word representation (word embeddings) within a neural network architecture..The main contribution of this paper is about word embeddings combination: several combination approaches are propos
shrill
发表于 2025-3-24 04:36:36
Semi-extractive Multi-document Summarization via Submodular Functions,ee monotone and submodular measures to detect importance of a sentence including Coverage, Relevance, and Compression. We apply an efficient scalable greedy algorithm to generate a summary which has a near optimal solution when its scoring functions are monotone nondecreasing and submodular. We use
maculated
发表于 2025-3-24 08:28:05
http://reply.papertrans.cn/88/8765/876449/876449_16.png
Compassionate
发表于 2025-3-24 13:07:13
Acoustical Frame Rate and Pronunciation Variant Statistics,rough a forced alignment of large amounts of speech signals with their possible pronunciations variants. Such alignments are usually performed using a 10 ms frame shift acoustical analysis. Therefore, the three emitting state structure of conventional acoustic hidden Markov models introduces a minim
公猪
发表于 2025-3-24 18:14:40
http://reply.papertrans.cn/88/8765/876449/876449_18.png
EXULT
发表于 2025-3-24 19:47:06
Automatic Detection of Voice Disorders, in the voice production can be caused by different diseases which can be treated better if they are diagnosed in an early stage. In this research we introduce a series of measurements based on continuous speech which could be a good base of developing a system that can automatically distinguish sev
CANE
发表于 2025-3-24 23:27:06
Semantic Features for Dialogue Act Recognition,tic information. The main goal of this paper is thus to propose semantic features and integrate them into a dialogue act recognition task to improve the recognition score. Three different feature computation approaches are proposed, evaluated and compared: Latent Dirichlet Allocation and the HAL and