Carlo Scagliola

Method of and device for testing the quality of digital speech‐transmission equipment

A device for analyzing the quality of digital speech-transmission equipment, specifically a speech coder, comprises generators of white-noise signals, sinusoidal signals, frequency-shaped signals and artificial speech-like signals connected to the coder and to an adaptive transversal filter in parallel therewith. The filter and the coder feed output signals to a subtractor which produces an error or noise signal by deducting digital samples of the filter output signal from corresponding samples of the coder output signal. The error signal is fed back to the filter during a first testing phase to control the periodic modification of weighting coefficients computed by a multiplicity of updating cells in the filter for multiplicative combination with incoming samples of the test signal. The first phase ends when the coefficients converge to fixed values representative of the linear characteristics of the coder and the error signal assumes a substantially constant near-zero level. In a second testing phase the error signal from the subtractor and the corrected output signal of the filter are fed to a quality analyzer which calculates one or more parameters from a set comprising a total signal-to-noise ratio, a simple segmental signal-to-noise ratio and a frequency-weighted segmental signal-to-noise ratio, the parameters being linearly combined to produce an integer between zero and ten indicative of the transmission quality of the coder.

Syntax driven recognition of connected words by Markov models

This paper describes a connected speech recognition system based on Markov models. The performance of this system was analyzed and compared with that of a system which uses prototypes of words instead of Markov models. Some preliminary results are reported with reference to the recognition of connected digits.

Continuous speech recognition via diphone spotting a preliminary implementation

The paper describes a preliminary implementation of a continuous speech recognition system based on the concept of diphone spotting. This consists of continuously measuring the similarity of the current portion of signal pattern with a complete set of selected phonetic events, called diphones because the most significant of them are transitions between pairs of phonemes. The entire set of measures feeds a linguistic decoder that operates on a state space representation of the language, whose states are the diphones that compose the words of the lexicon. Durational constraints and optional phonological rules are included in the language representation. The linguistic decoder recognizes the sentence as that path through the network which attains the highest cumulative similarity score. Additionally, the decoder has the task of detecting the end of the sentence. A preliminary test on 50 sequences of 3 to 7 connected digits gave recognition rates as high as 99.6% on digits, or 98% on sequences, thus confirming the validity of this approach.

Multiprocessor architecture for real-time speech recognition systems

In this paper we describe a multiprocessor architecture used in a real-time interactive system based on the recognition of sentences composed of sequences of isolated utterances. Anyway, the same general criteria could as well be used in the continuous speech framework. Particular attention is devoted to the control machanisms which supervise the system and, finally, detailed examples are given.

An identification method for objective quality measurements on speech waveform coders

Objective quality measures provide an economic and practic alternative to the tedious and expensive subjective tests. Furthermore some recently proposed measures predict quality values which correlate quite well with the subjective ones. In this paper the architecture of a measuring equipment to conduct objective quality measures on speech waveform coders is presented and discussed. Only the input and output signals from the coder are needed and an identification algorithm is used to separate the distortion effects introduced by the coder from other nondistorting effects as gains, time shifts, low-pass filtering and so on. The performance of the method is investigated under different conditions and the correlation with subjectively determined results is also reported for a number of ADPCM coders with largely varying parameters.

Real-time large vocabulary word recognition via diphone spotting and multiprocessor implementation

This paper describes Elsag’s Large Vocabulary Isolated Word Recognition system DSPELL. The system makes use of a diphone-based speech model and an extremely efficient word decoding algorithm, and is implemented on Elsag’s multiprocessor EMMA-2 1. DSPELL requires a very convenient training session and features a high recognition performance and real-time response on lexicons of up to 2,000 words.

The perception of spectrally shaped additive noise in speech

The spectrum of additive noise was systematically varied according to three variables: correspondance between the noise and speech spectrum (4 levels), equal weighting or weighting according to the width of the Articulation Index bands, SNR of the output signal (2 levels). All combinations of these 16 noise conditions were added to high quality speech sampled at 8kHz, and presented to listeners for judgment. Difference judgments were analyzed by a multidimensional scaling procedure to gain information about the number and nature of the important subjective variables. Overall quality ratings were obtained as a measure of the improvement in perceived quality due to the spectral shaping of the additive noise. Several objective measures were also tested for their sensitivity in predicting the overall quality ratings.