Learning of Word Boundaries In Continuous Speech Using Time Delay Neural Networks
نویسنده
چکیده
This paper presents early research results for a method for training neural networks to segment previously unseen continuous speech data into words, without the use of a lexicon or a speech recognition engine, or any other forms of supervision. The initial word segmentation is derived through a simple and naï ve method, and our method then leverages the time-invariant properties of the TDNN to derive better hypotheses, which is then used to recursively re-train the neural network.
منابع مشابه
A time-delay neural network architecture for isolated word recognition
-A translation-invariant back-propagation network is described that performs better than a soph&ticated continuous acoustic parameter hidden Markov model on a noisy, lO0-speaker confusable vocabulary isolated word recognition task. The network's replicated architecture permits it to extract precise information from unaligned training patterns selected by a naive segmentation rule. Keywords--Iso...
متن کاملIntegrating Time Alignment and Neural Networks for High Performance Continuous Speech Recognition
Successful application of existing connectionist methods to continuous speech recognition requires the use or time-alignment procedures. These procedures. usually based on dynamic programming, provide means for supervising the training of neural networks. This paper describes two systems in which neural network classifiers are merged with dynamic programming (DP) time alignment methods to produ...
متن کاملStable Rough Extreme Learning Machines for the Identification of Uncertain Continuous-Time Nonlinear Systems
Rough extreme learning machines (RELMs) are rough-neural networks with one hidden layer where the parameters between the inputs and hidden neurons are arbitrarily chosen and never updated. In this paper, we propose RELMs with a stable online learning algorithm for the identification of continuous-time nonlinear systems in the presence of noises and uncertainties, and we prove the global ...
متن کاملMulti-State Time Delay Neural Networks for Continuous Speech Recognition
Alex Waibel Carnegie Mellon University Pittsburgh, PA 15213 [email protected] We present the "Multi-State Time Delay Neural Network" (MS-TDNN) as an extension of the TDNN to robust word recognition. Unlike most other hybrid methods. the MS-TDNN embeds an alignment search procedure into the connectionist architecture. and allows for word level supervision. The resulting system has the ability to ma...
متن کاملFast Word Detection in a Speech Using New High Speed Time Delay Neural Networks
This paper presents a new approach to speed up the operation of time delay neural networks for fast detecting a word in a speech. The entire data are collected together in a long vector and then tested as a one input pattern. The proposed fast time delay neural networks (FTDNNs) use cross correlation in the frequency domain between the tested data and the input weights of neural networks. It is...
متن کامل