Cepstral normalisation and the signal to noise ratio spectrum in automatic speech recognition.
Related publications (85)
Graph Chatbot
Chat with Graph Search
Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.
DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.
In this paper, we consider the problem of speaker verification as a two-class object detection problem in computer vision, where the object instances are 1-D short-time spectral vectors obtained from the speech signal. More precisely, we investigate the ge ...
Recent high-density CMOS-based MEAs contain thousands of electrodes. Hence, an off-chip acquisition system performing data compression prior to visualization and analysis on a computer is required in order to handle the large amount of data generated by ac ...
The lanthanide binuclear helicate [Ln(2)(L-C2)(3)] has been embedded into bare and NH2-functionalized silica nanoparticles (NPs) using water-in-oil microemulsion technique. TEM analysis reveals both [Ln(2)(L-C2)(3)]@SiO2 and [Ln(2)(L-C2)(3)]@SiO2/NH2 nanop ...
The standard approach to speaker verification is to extract cepstral features from the speech spectrum and model them by generative or discriminative techniques. We propose a novel approach where a set of client-specific binary features carrying maximal di ...
When combined with cepstral normalisation techniques, the features normally used in Automatic Speech Recognition are based on Signal to Noise Ratio (SNR). We show that calculating SNR from the outset, rather than relying on cepstral normalisation to produc ...
In this paper, we consider the problem of speaker verification as a two-class object detection problem in computer vision, but the object instances are 1-D short-time spectral vectors obtained from the speech signal. More precisely, we investigate the gene ...
Using a transformation based at least in part on a non-simple orthogonal or unitary matrix, data may be transmitted over a data bus in a manner that is resilient to one or more types of signal noise, that does not require a common reference at the transmis ...
High spatial (~cm) and spectral (~MHz) resolution Brillouin sensing is realized with enhanced signal to noise ratio using a pre-activated acoustic field and an optical phase control over the interrogating pulse. Pre-activation of the acoustic field preserv ...
In this paper, we obtain a family of lower bounds for the sum capacity of code-division multiple-access (CDMA) channels assuming binary inputs and binary signature codes in the presence of additive noise with an arbitrary distribution. The envelope of this ...
Institute of Electrical and Electronics Engineers2009
When combined with cepstral normalisation techniques, the features normally used in Automatic Speech Recognition are based on Signal to Noise Ratio (SNR). We show that calculating SNR from the outset, rather than relying on cepstral normalisation to produc ...