Automatic speech recognition over error-prone wireless networks

Tan, Zheng‐Hua; Dalsgaard, Paul; Lindberg, Børge

doi:10.1016/j.specom.2005.05.007

Cited by 42 publications

(23 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DSR systems offer some advantages over other architectures. Recent comparative studies have shown the superior performance of DSR to codec-based ASR [4]. However, in a DSR system, transmission errors in the form of random packet loss and packet burst loss still need to be taken into consideration.…”

Section: Dsr Architecture and Standardsmentioning

confidence: 99%

“…Error-robustness techniques are categorised in [4] under the headings client-based error recovery, and server-based error concealment. Client-based techniques include retransmission, interleaving and forward error correction (FEC).…”

Section: Channel Models and Loss Compensationmentioning

confidence: 99%

“…Server-based methods include feature reconstruction, by means of repetition or interpolation, and error correction in the ASR-decoding stage. Reference [4] provides a survey of robustness issues related to network degradations and presents a number of analyses and experiments with a focus on transmission error robustness.…”

Section: Channel Models and Loss Compensationmentioning

confidence: 99%

See 2 more Smart Citations

Robust Distributed Speech Recognition Using Auditory Modelling

Flynn¹,

Jones²

2012

Modern Speech Recognition Approaches With Case Studies

View full text Add to dashboard Cite

Section: Dsr Architecture and Standardsmentioning

confidence: 99%

Section: Channel Models and Loss Compensationmentioning

confidence: 99%

Section: Channel Models and Loss Compensationmentioning

confidence: 99%

See 1 more Smart Citation

Robust Distributed Speech Recognition Using Auditory Modelling

Flynn¹,

Jones²

2012

Modern Speech Recognition Approaches With Case Studies

View full text Add to dashboard Cite

“…In the client-server distributed speech recognition (DSR) system architecture, speech features suffering from corruption or frame loss are inevitable when DSR is used over error prone packet-switched networks (Tan et al, 2005). Considerable efforts have been devoted to alleviate the performance degradation due to corruption or frame loss in DSR.…”

Section: Introductionmentioning

confidence: 99%

Model adaptation method for recognition of speech with missing frames

Lee

Jean

2014

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

In distributed speech recognition (DSR), data packets may be lost over error prone channels. A commonly used approach to rectify this is to reconstruct a full frame rate data sequence for recognition using linear interpolation. In this study, an error-concealment decoding method that dynamically adapts the transition probabilities of hidden Markov models to match the frame loss observation sequence is proposed. Experimental results show that a DSR system using the proposed method can achieve the same level of accuracy as a data reconstruction method, is more robust against heavy frame loss, and significantly reduces the computation time.

show abstract

“…At the beginning of the use of mobile phones, speech recognition was only feasible as "distributed speech recognition" (DSR) whose idea was to perform all the CPU intensive calculations in central mainframes while only the speech capture and feature extraction ran in the mobile terminal. Wireless networks presented several problems that posed challenges to the performance of ASR systems: bandwidth constraints and transmission errors [4]. Recently the terminals have gained enough CPU power so that the idea of distributed recognition has declined and some laboratories are even introducing complex stochastic predictive algorithms in the terminal to reduce error rates [5].…”

Section: Introductionmentioning

confidence: 99%

A speech interface for air traffic control terminals

Ferreiros

Pardo

Córdoba

et al. 2012

Aerospace Science and Technology

View full text Add to dashboard Cite

Abstract-Several issues concerning the current use of speech interfaces are discussed and the design and development of a speech interface that enables air traffic controllers to command and control their terminals by voice is presented. A special emphasis is made in the comparison between laboratory experiments and field experiments in which a set of ergonomics-related effects are detected that can not be observed in the controlled laboratory experiments.The paper presents both objective and subjective performance obtained in field evaluation of the system with student controllers at an air traffic control (ATC) training facility. The system exhibits high word recognition test rates (0.4% error in Spanish and 1.5% in English) and low command error (6% error in Spanish and 10.6% error in English in the field tests). Subjective impression has also been positive, encouraging future development and integration phases in the Spanish ATC terminals designed by Aeropuertos Españoles y Navegación Aérea (AENA).

show abstract

Automatic speech recognition over error-prone wireless networks

Cited by 42 publications

References 63 publications

Robust Distributed Speech Recognition Using Auditory Modelling

Robust Distributed Speech Recognition Using Auditory Modelling

Model adaptation method for recognition of speech with missing frames

A speech interface for air traffic control terminals

Contact Info

Product

Resources

About