Spoken Word Lists Research Articles

Speech Recognition in Natural Background Noise

In the real world, human speech recognition nearly always involves listening in background noise. The impact of such noise on speech signals and on intelligibility performance increases with the separation of the listener from the speaker. The present behavioral experiment provides an overview of the effects of such acoustic disturbances on speech perception in conditions approaching ecologically valid contexts. We analysed the intelligibility loss in spoken word lists with increasing listener-to-speaker distance in a typical low-level natural background noise. The noise was combined with the simple spherical amplitude attenuation due to distance, basically changing the signal-to-noise ratio (SNR). Therefore, our study draws attention to some of the most basic environmental constraints that have pervaded spoken communication throughout human history. We evaluated the ability of native French participants to recognize French monosyllabic words (spoken at 65.3 dB(A), reference at 1 meter) at distances between 11 to 33 meters, which corresponded to the SNRs most revealing of the progressive effect of the selected natural noise (−8.8 dB to −18.4 dB). Our results showed that in such conditions, identity of vowels is mostly preserved, with the striking peculiarity of the absence of confusion in vowels. The results also confirmed the functional role of consonants during lexical identification. The extensive analysis of recognition scores, confusion patterns and associated acoustic cues revealed that sonorant, sibilant and burst properties were the most important parameters influencing phoneme recognition. . Altogether these analyses allowed us to extract a resistance scale from consonant recognition scores. We also identified specific perceptual consonant confusion groups depending of the place in the words (onset vs. coda). Finally our data suggested that listeners may access some acoustic cues of the CV transition, opening interesting perspectives for future studies.

Acoustic masking disrupts time-dependent mechanisms of memory encoding in word-list recall.

Recall of recently heard words is affected by the clarity of presentation: Even if all words are presented with sufficient clarity for successful recognition, those that are more difficult to hear are less likely to be recalled. Such a result demonstrates that memory processing depends on more than whether a word is simply "recognized" versus "not recognized." More surprising is that, when a single item in a list of spoken words is acoustically masked, prior words that were heard with full clarity are also less likely to be recalled. To account for such a phenomenon, we developed the linking-by-active-maintenance model (LAMM). This computational model of perception and encoding predicts that these effects will be time dependent. Here we challenged our model by investigating whether and how the impact of acoustic masking on memory depends on presentation rate. We found that a slower presentation rate causes a more disruptive impact of stimulus degradation on prior, clearly heard words than does a fast rate. These results are unexpected according to prior theories of effortful listening, but we demonstrated that they can be accounted for by LAMM.

Speech Intelligibility and Recall of Spoken Material Heard at Different Signal‐to‐noise Ratios and the Role Played by Working Memory Capacity

SummaryWe studied speech intelligibility and memory performance for speech material heard under different signal‐to‐noise (S/N) ratios. Pre‐experimental measures of working memory capacity (WMC) were taken to explore individual susceptibility to the disruptive effects of noise. Thirty‐five participants first completed a WMC‐operation span task in quiet and later listened to spoken word lists containing 11 one‐syllable phonetically balanced words presented at four different S/N ratios (+12, +9, +6, and +3). Participants repeated each word aloud immediately after its presentation, to establish speech intelligibility and later on performed a free recall task for those words. The speech intelligibility function decreased linearly with increasing S/N levels for both the high‐WMC and low‐WMC groups. However, only the low‐WMC group had decreasing memory performance with increasing S/N levels. The memory of the high‐WMC individuals was not affected by increased S/N levels. Our results suggest that individual differences in WMC counteract some of the negative effects of speech noise. Copyright © 2012 John Wiley & Sons, Ltd.

Effects of talker, rate, and amplitude variation on recognition memory for spoken words.

This study investigated the encoding of the surface form of spoken words using a continuous recognition memory task. The purpose was to compare and contrast three sources of stimulus variability--talker, speaking rate, and overall amplitude--to determine the extent to which each source of variability is retained in episodic memory. In Experiment 1, listeners judged whether each word in a list of spoken words was "old" (had occurred previously in the list) or "new." Listeners were more accurate at recognizing a word as old if it was repeated by the same talker and at the same speaking rate; however, there was no recognition advantage for words repeated at the same overall amplitude. In Experiment 2, listeners were first asked to judge whether each word was old or new, as before, and then they had to explicitly judge whether it was repeated by the same talker, at the same rate, or at the same amplitude. On the first task, listeners again showed an advantage in recognition memory for words repeated by the same talker and at same speaking rate, but no advantage occurred for the amplitude condition. However, in all three conditions, listeners were able to explicitly detect whether an old word was repeated by the same talker, at the same rate, or at the same amplitude. These data suggest that although information about all three properties of spoken words is encoded and retained in memory, each source of stimulus variation differs in the extent to which it affects episodic memory for spoken words.

Taking Charge of Remembering: Locus of Control and Older Adults' Memory for Speech

The relation between control beliefs and recall of spoken word lists and prose passages was assessed for 32 older adults, ages 62 to 85, in a task where they were given control over presentation of stimuli. They differed in the degree to which they believed that factors within their control (internals) or outside their control (externals) affected their intellectual functioning; they were similar in age, education, vocabulary, and digit span. They were required to stop the speech input at points of their own choosing to recall the stimuli on a segment-by-segment basis. Externals were more likely than internuls to make inaccurate predictions of the number of words they could remember and to choose longer segments than they could recall. Results suggest that externals are poorer than internals in monitoring on-line memory processing.

Sources of variability as linguistically relevant aspects of speech

Recent research has suggested that only linguistically relevant sources of variability, or those which affect the perception of spoken words, are retained in long-term memory. The present study sought to determine if the linguistic relevance of surface characteristics such as speaking rate, overall amplitude, and vocal effort would differentially affect memory retention. For each source of variability, a continuous recognition memory task was used in which listeners were presented with a list of spoken words and asked to judge whether each word in the list was ‘‘old’’ (had occurred previously on the list) or ‘‘new.’’ Results showed that listeners were better able to identify a word as ‘‘old’’ if the word was repeated at the same speaking rate (condition 1), overall amplitude (condition 2), or vocal effort (condition 3), suggesting that the individual surface characteristics which comprise each type of variability were encoded into memory. However, speaking rate and vocal effort produced a greater effect on recognition memory than overall amplitude, indicating that all sources of variability may not be processed and encoded to the same extent. Rather, memory for surface characteristics may be related to the amount of linguistically relevant information each source of variability contains.

The effect of talker rate and amplitude variation on memory representation of spoken words.

This study investigated the encoding of spoken word attributes in a task of continuous recognition memory. In experiment 1, subjects were presented with a list of spoken words, and for each word they judged whether the word was ‘‘old’’ (had occurred previously in the list) or ‘‘new.’’ Results showed that subjects were more accurate at recognizing a word as ‘‘old’’ if it was repeated in the same voice (condition 1), and at the same rate of speech (condition 2); however, no recognition advantage was found for words repeated at the same amplitude (condition 3). In experiment 2, subjects gave an additional explicit judgment as to whether ‘‘old’’ words were repeated in the same voice (condition 1), rate (condition 2), or amplitude (condition 3). Subjects again showed a recognition advantage for words repeated in the same voice and rate, but no advantage occurred in the amplitude condition. However, subjects in all three conditions were significantly above chance accuracy in recognizing whether an ‘‘old’’ word was repeated in the same voice, rate or amplitude. These data suggest that, even though voice, rate, and amplitude information may be encoded differently, information along all three dimensions is retained in memory. [Work supported by NIH.]

Technical advances in the analysis of single, acoustically evoked potentials

A method that evaluates the degree and, perhaps, the form and time position of single EEG evoked potentials is described. Its sensitivity and possible error sources are estimated. The method uses a double summation of cross products between a template and the EEG. This double summation is repeated at 25 msec steps through the EEG creating a correlogram. The analysis is given for the template times itself ) i.e., optimal matching), for the template buried in the specified noise of a 2 c/sec wave, the template buried in the unspecified noise of an EEG, and, for an EEG stimulated by a list of spoken words.

Audio‐visual aids

ARGUMENTATION AND DEBATE. Dual track tape recording. 33/4 IPS. Two hour tape by Dr. James H. McBath. Cost: $14.95. Available through Campus World, Inc., 3757 Wilshire Boulevard, Los Angeles 5, California. COMMUNICATION. 16 mm. film. 10 minutes. Sound. Color and Black and White. Purchase: Color $110.00; Black and White $55.00. Rental: $5.50. One of the Primary Modern Civilization Series. A Camerart Picture produced and narrated by John F. Criswell. Available through Gateway Productions, Inc., 1859 Powell Street, San Francisco 11, California. PATRICK HENRY'S FAMOUS SPEECH and THE BILL OF RIGHTS; CONSTITUTIONAL AMENDMENTS—13th, 14th, 15th and JEFFERSON'S FIRST INAUGURAL ADDRESS. Documentary Enrichments Records. 12 inch records with 15–20 minutes on each side. 331/3 RPM. Cost: $5.29 each. Available through Enrichment Teaching Materials, 246 Fifth Avenue, New York 1, New York. GENERAL ELECTRIC ETV OPERATING CENTER. A Complete Audio‐Visual Control Package. For detailed information contact Educational Television Sales, Technical Products Operation, General Electric Company, 212 West Division Street, Syracuse 3, New York. FOREIGN LANGUAGES ANIMATED TRANSPARENCY. A device to aid in the teaching of the position of the articulators in forming speech sounds. For further information contact Ben Lorenz Associates, 200 Park Avenue, New York, New York. SPEECHPHONE COURSES. (1. American Speech Sounds and Rhythm—Elementary; 2. American Speech Sounds and Rhythm—Intermediate; 3. American Speech Sounds and Rhythm—Advanced; 4. Spoken Word List). Three 12 inch records per course. 331/3 RPM. Created by Hazel P. Brown. Issued by Speech‐phone Record Division. Records produced by Columbia Recording Studios. First issued 1950; third edition, 1959. Cost: $29.50 per course. Available through Speechphone Record and Tape Division, Box 425, Grand Central Station, New York 17, New York. DIALECTS. 16 mm. film. 30 minutes. Sound. Black and White. Number 6 in a series of 13 films of the National Educational Television (NET) films on LANGUAGE AND LINGUISTICS by Dr. Henry Lee Smith, Jr. made at the University of Buffalo, New York, for the Metropolitan Television Center of New York City. Purchase: $125.00. Rental (five days): $5.25 plus postage. Available through Association Films, 347 Madison Avenue, New York 17, New York, or NET Film Service, Audio Visual Center, Bloomington, Indiana. DIRECTING A PLAY. A Lecture by Tyrone Guthrie. 331/3 RPM. 12 inch record. Album number: FL 9840. Cost: $5.95. Available through Folkways Records and Service Corporation, 121 West 47th Street, New York 36, New York. ELEMENTARY SET DESIGN. Series of five 35mm. film strips. Color. Written and directed by Dr. Wendell Cole, Stanford University; Editorial Consultant, Dr. James Butler, University of Southern California; Original Art Work, Keith Cooledge. Cost: $2750; $6.50 each strip. Available through Comma, 1535 Ivar Avenue, Hollywood 28, California. This series of five film strips is excellently done, has exciting pictures for illustrations, and should be useful for a number of different purposes. THE SECOND SHEPHERD'S PLAY. Directed by Howard Sackler. 12 inch record. 331/3 RPM. Playing time approximately 40 minutes. Album number: TC1032. Cost: $5.95. Available from Caedmon Records, 461 8th Avenue, New York 1, New York. GEORGE JEAN NATHAN'S THE NEW AMERICAN CREDO. Read by Julie Haydon Nathan. 1962. 331/3 RPM. Album number: FL 9760. Cost: $5.95. Available through Folkways Records and Service Corporation, 121 West 47th Street, New York. POETRY IN THE ROUND. A Poetry Workshop. Narration and poems by George Abbe. 12 inch record. 331/3 RPM. Album number: FI 9164. Cost: $5.95‐ Available through Folkways Records and Service Corporation, 121 West 47th Street, New York, New York.