Dictionary learning: performance through consistency
نویسنده
چکیده
We present rst results from our e orts in automatically increasing and adapting phonetic dictionaries for spontaneous speech recognition. Spontaneous speech adds a variety of phenomena to a speech recognition task: false starts [1], human and nonhuman noises [2], new words [3] and alternative pronunciations. All of these phenomena have to be tackled when adapting a speech recognition system for spontaneous speech. For phonetic dictionaries (especially for spontaneous speech) it is important to choose the pronunciations of a word according to the frequency in which they appear in the database rather than the \correct" pronunciation as it might be found in a lexicon. Additionally modi cations of the dictionary should not lead to a higher phoneme confusability. Therefore we propose a data-driven approach to add new pronunciations to a given phonetic dictionary, in a way that they model the given occurrences of words in the database. We show how even a simple approach can lead to signi cant improvements in recognition performance. First experiments have been performed on the German Spontaneous Scheduling Task (GSST), using the speech recognition engine of JANUS-2 [4, 5, 6], the spontaneous speech-to-speech translation system of the Interactive Systems Laboratories at Carnegie Mellon and Karlsruhe University.
منابع مشابه
Structured Low-rank Coding for Top-down Saliency
In this paper, we propose a structured low-rank coding method for top-down saliency detection. Both spatial consistency and structured information are considered in our proposed method. Spatial consistency encourages local image patches which are spatially close in an image to have similar representations. Structured information facilitates the patches from the target to have similar representa...
متن کاملA Novel Face Detection Method Based on Over-complete Incoherent Dictionary Learning
In this paper, face detection problem is considered using the concepts of compressive sensing technique. This technique includes dictionary learning procedure and sparse coding method to represent the structural content of input images. In the proposed method, dictionaries are learned in such a way that the trained models have the least degree of coherence to each other. The novelty of the prop...
متن کاملSpeech Enhancement using Adaptive Data-Based Dictionary Learning
In this paper, a speech enhancement method based on sparse representation of data frames has been presented. Speech enhancement is one of the most applicable areas in different signal processing fields. The objective of a speech enhancement system is improvement of either intelligibility or quality of the speech signals. This process is carried out using the speech signal processing techniques ...
متن کاملA New Method for Speech Enhancement Based on Incoherent Model Learning in Wavelet Transform Domain
Quality of speech signal significantly reduces in the presence of environmental noise signals and leads to the imperfect performance of hearing aid devices, automatic speech recognition systems, and mobile phones. In this paper, the single channel speech enhancement of the corrupted signals by the additive noise signals is considered. A dictionary-based algorithm is proposed to train the speech...
متن کاملEffects of Entrepreneurial Orientation on the Export of Small and Medium Companies through the mediation of Organizational Learning capability and Innovation Performance
In the current world, export development is not limited to increasing foreign exchange earnings through exports of goods and services, but also plays a more important role as a growth strategy and even economic development in the world. Therefore, how to become a successful exporter is becoming more and more important. The main objective of this study was to study the effect of entrepreneurial ...
متن کامل