Telephone Band Research Articles

Speech intelligibility and talker identification with non-telephone frequencies.

Although the telephone band (0.3-3 kHz) provides sufficient information for speech recognition, the contribution of the non-telephone band (<0.3 and >3 kHz) is unclear. To investigate its contribution, speech intelligibility and talker identification were evaluated using consonants, vowels, and sentences. The non-telephone band produced relatively good intelligibility for consonants (76.0%) and sentences (77.4%), but not vowels (11.5%). The non-telephone band supported good talker identification only with sentences (74.5%), but not vowels (45.8%) or consonants (10.8%). Furthermore, the non-telephone band cannot produce satisfactory speech intelligibility in noise at the sentence level, suggesting the importance of full-band access in realistic listening.

The distribution of speaker information in Dutch fricatives /s/ and /x/ from telephone dialogues.

Although previous work has shown that some speech sounds are more speaker-specific than others, not much is known about the speaker information of the same segment in different linguistic contexts. The present study, therefore, investigated whether Dutch fricatives /s/ and /x/ from telephone dialogues contain differential speaker information as a function of syllabic position and labial co-articulation. These linguistic effects, established in earlier work on read broadband speech, were first investigated. Using a corpus of Dutch telephone speech, results showed that the telephone bandwidth captures the expected effects of perseverative and anticipatory labialization for dorsal fricative /x/, for which spectral peaks fall within the telephone band, but not for coronal fricative /s/, for which the spectral peak falls outside the telephone band. Multinomial logistic regression shows that /s/ contains slightly more speaker information than /x/ in telephone speech and that speaker information is distributed across the speech signal in a systematic way; even though differences in classification accuracy were small, codas and tokens with labial neighbors yielded higher scores than onsets and tokens with non-labial neighbors for both /s/ and /x/. These findings indicate that speaker information in the same speech sound is not the same across linguistic contexts.

Bandwidth Extension of Speech Signals: A Comprehensive Review

Telephone systems commonly transmit narrowband (NB) speech with an audio bandwidth limited to the traditional telephone band of 300-3400 Hz. To improve the quality and intelligibility of speech degraded by narrow bandwidth, researchers have tried to standardize the telephonic networks by introducing wideband (50-7000 Hz) speech codecs. Wideband (WB) speech transmission requires the transmission network and terminal devices at both ends to be upgraded to the wideband that turns out to be time-consuming. In this situation, novel Bandwidth extension (BWE) techniques have been developed to overcome the limitations of NB speech. This paper discusses the basic principles, realization, and applications of BWE. Challenges and limitations of BWE are also addressed.

Spectral shifting of speaker‐specific information for narrow band telephonic speaker recognition

Speech-based authentication system can perform remote authentication over telephone channels. However, telephone channels are restricted to a bandwidth of ∼0–4 kHz while studies on the distribution of speaker-specific information in the speech spectrum strongly suggests that useful speaker-specific information is present above 4 kHz. A method to shift a part of this speaker-specific information above 4 kHz into the telephone bandwidth in place of less speaker-specific information originally present below 4 kHz is proposed. Speaker recognition experiments conducted using the proposed method leads to ∼18.5% relative improvement on equal error rate when compared to a system using the conventional telephone band speech, as evaluated on the Intelligence Advanced Research Projects Activity (IARPA) Babel Program Tamil language collection.

Multiband shorted monopole antenna

A new type of multiband shark-fin antenna printed on 1.6 mm thick FR4 for automotive applications is presented. The antenna is mounted over a 1 m × 1 m-metal ground plane which represents a car’s roof. The overall size of the antenna is 57 mm × 40 mm × 1.6 mm which is λ/6.5 × λ/9 × λ/231.5 at the lowest frequency of operation of 0.810 GHz. The antenna is in the form of a single-sided PCB and has a measured gain of 0–6.7 dBi over the entire range of 0.810–5.84 GHz covering the mobile telephone bands (GSM, DCS and PCS), both GPS bands and Bluetooth, WiFi and V2X bands on −6 dB measured reflection coefficient criteria. The measured co-polar radiation patterns in the azimuth (XZ-plane) and elevation (YZ-plane) planes depict omnidirectional radiation behaviour.

Tunable Multiband Handset Antenna Operating at VHF and UHF Bands

A compact multiband tunable PIFA system using a varactor as an active tuning component is presented covering 88-2175 MHz. The antenna was designed to cover the mobile telephone bands (GSM850, 900, 1800, 1900, UMTS) by controlling the value of the capacitance across the gap in the slotted PIFA and two broadcast bands, FM radio (76-108 MHz) & mobile television DVB-H (470-702 MHz), were added by designing an appropriate matching circuit into the system. The antenna measured λ/15 × λ/43 × λ/19 at 470 MHz and was easily integrated into a small mobile phone handset. Simulations are verified by measurements.

Evaluation of an Artificial Speech Bandwidth Extension Method in Three Languages

Quality and intelligibility of narrowband telephone speech can be improved by artificial bandwidth extension (ABE), which extends the speech bandwidth using only information available in the narrowband speech signal. This paper reports a three-language evaluation of an ABE method that has recently been launched in several of Nokia's mobile telephone models. The method extends the speech bandwidth to frequencies above the telephone band by first utilizing spectral folding and then modifying the magnitude spectrum of the extension band with spline curves. The performance of the method was evaluated by formal listening tests in American English, Russian, and Mandarin Chinese. The results of the listening tests indicate that ABE processing improved the subjective quality of coded narrowband speech in all these languages. Differences between bandwidth-extended American English test sentences and their original wideband counterparts were also evaluated using both an objective distance measure that simulates the characteristics of human hearing and a conventional spectral distortion measure. The average objective error was calculated for different categories of speech sounds. The error was found to be smallest in nasals and semivowels and largest in fricative sounds.

Design of a Shape-Constrained Dual-Band Polygonal Monopole for Car Roof Mounting

This paper describes the design of a new compact dual-band roof antenna (which is capable of operating in both U.S. and European mobile telephone bands) for automotive applications. The flipper shape of the antenna and its dimensional constrains are chosen according to prevailing automotive market trends. The chosen basic geometry of the radiating element is a planar printed monopole structured as a vertical fan with two narrow strips folding toward the ground plane and etched on a low-cost, medium-permittivity thin substrate. The geometrical variations of the basic radiating structure are introduced afterward to constrain the antenna shape under the assigned profile and to allow easy integration of a small Global Positioning System patch antenna without affecting radiation and bandwidth performances. The integration of a matching network directly over the etched monopole trace allows a good input matching over the lower and upper operative bands to be obtained, slightly affecting the radiation efficiency of the whole antenna.

Neural Network-Based Artificial Bandwidth Expansion of Speech

The limited bandwidth of 0.3-3.4 kHz in current telephone systems reduces both the quality and the intelligibility of speech. Artificial bandwidth expansion is a method that expands the bandwidth of the narrowband speech signal in the receiving end of the transmission link by adding new frequency components to the higher frequencies, i.e., up to 8 kHz. In this paper, a new method for artificial bandwidth expansion, termed Neuroevolution Artificial Bandwidth Expansion (NEABE) is proposed. The method uses spectral folding to create the initial spectral components above the telephone band. The spectral envelope is then shaped in the frequency domain, based on a set of parameters given by a neural network. Subjective listening tests were used to evaluate the performance of the proposed algorithm, and the results showed that NEABE speech was preferred over narrowband speech in about 80% of the test cases

Multiband PIFA Vehicle Telematics Antennas

Compact, low-cost antennas for multiband telematics applications on vehicles are described. The multifrequency band antennas are variations on the printed inverted-F antenna (PIFA) principle. One type of PIFA is a compact variation of a conventional antenna while the other is a novel planar version customized for different mounting positions such as bumpers, window, and roof mount. In both derivations the dimensions are small and have been specifically designed for low-cost production. Each antenna was designed for mobile telephone band coverage from 890 to 2100 MHz. Results from free-space simulations are included but more work needs to be carried out as the vehicles have a significant effect on antenna performance. On-vehicle tests are described where the antennas were placed on glass and plastic areas and integrated into an existing multipurpose roof mount antenna.

Multiband PCB antenna

A reduced height automotive antenna covering the three mobile telephone bands used around the world is described. The antenna is much more compact than a conventional monopole and is printed on a PCB for low-cost production. Circuit components are used to tune the antenna to cover the regional band variations for mobile telephone systems, enabling one design to cover them all. A voltage standing-wave ratio of better than 2 : 1 is achieved. Performance when used as a roof mount automotive antenna is close to that of a quarter-wave monopole antenna.

Hidden antennas for vehicles

The introduction of new telematics and broadcast systems into vehicles has led to a requirement for multiple antennas that can be hidden from view. This paper commences by presenting the results of simulations to identify the components of a car's structure that influence the radiation pattern of a printed VHF antenna on the rear windscreen. Two dual-band antenna designs are then presented for operation in the 900 and 1800 MHz telephone bands. The first is a planar inverted-F antenna that can be concealed in the bumpers, the second a hybrid structure based on the top-loaded monopole principle and mounted beneath the vehicle's roof.

Signal modeling for high-performance robust isolated word recognition

This paper describes speech signal modeling techniques which are well-suited to high performance and robust isolated word recognition. We present new techniques for incorporating spectral/temporal information as a function of the temporal position within each word. In particular, spectral/temporal parameters are computed using both variable length blocks with a variable spacing between blocks. We tested features computed with these methods using an alphabet recognition task based on the ISOLET database. The hidden Markov model toolkit (HTK) was used to implement the isolated word recognizer with whole word HMM models. The best accuracy achieved for speaker independent alphabet recognition, using 50 features, was 97.9%, which represents a new benchmark for this task. We also tested these methods with deliberate signal degradation using additive Gaussian noise and telephone band limiting and found that the recognition degrades gracefully and to a smaller degree than for control cases based on MFCC coefficients and delta cepstra terms.

Sound solutions for cellular phones

The ongoing miniaturization of cellular phones during the last years has led to the development of smaller and smaller speakers for these products. Today’s size of a typical speaker for GSM and CDMA phones is of a diameter of 15 and 13 mm, and we are about to see even smaller speakers on the market. In the design and type approval of the receiving part of the cellular phone, the reference is a standardized measuring ear, which seals toward the ear cap plane of the phone. Under these conditions, it is possible to achieve a flat frequency response representing the whole telephone band, even with speakers in small sizes. However, such a tiny speaker in combination with a very small cellular phone having a poor surface of contact to the human ear leads to heavily reduced speech quality, which cannot be reflected in the measurement result. In other words, there is a very poor correlation between the type real life telephone user situation. This presentation gives the guidelines on how to design smaller cellular phones prepared for type approval using small speakers and at the same time improve sound quality.

Miniature HTS spiral cavity resonator for mobile telephone bands

Work on a miniaturized cavity resonator operating at around 900 MHz is reported. This uses a superconducting spiral resonator to enable its size to be reduced substantially in comparison with conventional cavity realizations: the new cavity fits within a 2 cm cube. The superconducting spiral structure was made of YBa2Cu3Ox or Tl2Ba2CaCu2Ox on a 1 cm2 substrate of lanthanum aluminate (LaAlO3). The unloaded Q-factors at temperatures below Tc were ten and four times greater than with an equivalent copper spiral, for YBCO and TBCCO respectively. The reason for the poor result with TBCCO is presumed to be due to low film quality. The Q factors showed a relatively small power dependence from −40 dBm to +11 dBm input power. A simulation of the cavity predicted resonance at 860 MHz, compared with 840 MHz measured, and a Q-factor similar to that found with YBCO (≈ 1000).

Line-of-sight angle of arrival estimation in the outdoor multipath environment

Angle of arrival (AOA) estimation utilizing the MUltiple Signal Identification and Classification (MUSIC) algorithm is investigated for land vehicle location systems. A virtual antenna array is used to overcome problems inherent to MUSIC when operating in a highly correlated signal environment. Based on assumptions regarding the multipath radio environment, triangulation is possible by estimating the AOA of the line-of-sight (LOS) component in the received signal. Computer simulations are used to analyze the ability of techniques incorporating MUSIC to achieve this. Since triangulation is achieved using a narrowband signal, no new spectrum is required. Instead, the current cellular telephone bands as well as the existing basestation hardware could be exploited resulting in a cost-efficient and easily implemented system. A further cost reduction may be realized by processing received signal data while travelling along a nonlinear trajectory. In this way, an unambiguous angular spectrum is obtained with one antenna.

Remote stethoscope signal processing system

The present invention relates to a remote stethoscope system that allows a doctor at one location to listen to the stethoscope sounds coming from a stethoscope being used by a patient in his or her remotely-located home. The acoustic stethoscope sounds are converted into electrical signals, frequency shifted up to the telephone band, and then conveyed over a conventional telephone line. At the doctor's location, the signals are shifted down to their original frequencies and then converted back to audible sound for the doctor's analysis. The sensor used by the patient picks up a wide range of frequencies without any manipulation by the patient.

Coding methods for telephone band width speech

Coding methods for telephone band width speech

Using auditory models to enhance telephone speech for the hearing impaired

Previously the effectiveness of frequency shaping and compression achieved via manipulation of the short-term spectrum of telephone speech for mild-to-moderate hearing impaired individuals was presented [Terry etal., J. Acoust. Soc. Am. 88, S169 (1990)]. This paper compares this technique with two analysis/resynthesis techniques designed to enhance the consonant parts of speech. In one system a time-domain auditory model (similar to the model proposed Goldstein [Hear. Res. 49, 39–60 (1990)] was used to analyze the speech. A set of consonant identifying features was then used to drive the resynthesis of the speech from the periodicity spectrum. In the other system the speech was resynthesized from LPC coefficients obtained via LPC analysis of the telephone speech signal. Again a set of consonant identifying features derived from LPC analysis was used to drive the resynthesis. The consonant-enhanced telephone band speech was presented to subjects with sensori-neural hearing loss. A consonant-confusion analysis will be presented comparing and evaluating both analysis/resynthesis techniques with a frequency shaping technique based on the subjects audiogram. [Work supported by US West.]

Processing the telephone speech signal for the hearing impaired.

Speech intelligibility scores from 16 subjects with sensorineural hearing loss were evaluated using a digitized version of the California Consonant Test that was presented via headphones through a 300 to 3000 Hz bandpass filter to simulate the telephone band. Each subject was tested with an unprocessed signal that was frequency-equalized to compensate for the individual's hearing loss, and a signal that was equalized and compressed by the use of a compressor compression technique. Subjects were tested at three sound pressure levels above a pure-tone average threshold for frequencies 1 and 2 kHz. Two digital signal processing techniques designed to compensate for high-frequency hearing loss were examined: frequency domain processing and time domain processing. Frequency domain involved modification of the short-term spectrum obtained through a fast Fourier transform, whereas time domain processing involved passing the signal through a bank of finite impulse response filters. Both techniques showed significant intelligibility improvements (15-30%). In a second experiment, 16 additional subjects with high-frequency hearing loss compared an amplified telephone signal to three processed signals: (1) 6 dB per octave emphasis; (2) a signal frequency equalized for their hearing loss; and (3) a signal that was equalized for their hearing loss and was compressed according to their uncomfortable loudness levels. Most subjects preferred the signal with the 6 dB per octave emphasis.