Strategies for name recognition in automatic directory assistance systems

Schramm, Hauke; Rueber, B.; Kellner, Aaron

doi:10.1016/s0167-6393(99)00066-7

Cited by 25 publications

(18 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Within the same Turbo framework, it remains important to investigate (1) different schemes for parameter tuning, (2) other methods for exchanging bias information between different domains, as well as (3) extensions of this algorithm to more general set-ups. The algorithm is also expected to be incorporated into a spoken dialogue system for automatically acquiring new words.…”

Section: Wordmentioning

confidence: 99%

“…Such systems often suffer from dialogue breakdown at critical points that convey important information such as named entities or geographical locations. One successful approach proposed for error recovery in dialogue systems lies in speak-and-spell models, that prompt the user for the spelling of an unrecognized word [2,3]. In such cases, both the spoken spelling and word are available.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A turbo-style algorithm for lexical baseforms estimation

Choueiter

Ohannessian

Seneff

et al. 2008

2008 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

In this research, an iterative and unsupervised Turbo-style algorithm is presented and implemented for the task of automatic lexical acquisition. The algorithm makes use of spoken examples of both spellings and words and fuses information from letter and subword recognizers to boost the overall lexical learning performance. The algorithm is tested on a challenging lexicon of restaurant and street names and evaluated in terms of spelling accuracy and letter error rate. Absolute improvements of 7.2% and 3% (15.5% relative improvement) are obtained in the spelling accuracy and the letter error rate respectively following only 2 iterations of the algorithm.

show abstract

Section: Wordmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A turbo-style algorithm for lexical baseforms estimation

Choueiter

Ohannessian

Seneff

et al. 2008

2008 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

show abstract

“…There are many potential applications, including automated directory assistance (Lehtinen et al, 2000;San-Segundo et al, 2002;Schramm et al, 2000) and the identification of city names for travel services (Lamel et al, 2000).…”

Section: Introductionmentioning

confidence: 99%

“…Again, the best result was obtained when the spelling was weighted more heavily than the fluently spoken utterance (98% accuracy). Schramm et al (2000) explored a similar method, although in this case equal weighting was given to both the spoken and spelled hypotheses. Similar levels of accuracy (92.5% first-best and 97.3% three-best) were obtained on a large inventory of names (approximately 190 000).…”

Section: Introductionmentioning

confidence: 99%

Usability of dialogue design strategies for automated surname capture

Davidson

McInnes

Jack

2004

Speech Communication

View full text Add to dashboard Cite

Surname capture via automatic speech recognition over the telephone has many commercial applications, including automated directory assistance and travel reservation services. This paper presents a usability evaluation of three different dialogue designs for automated surname capture, within the context of a flight reservation service. The three designs explored were: a Speak Only strategy, in which callers simply say the surname; a One Stage Speak and Spell strategy in which callers speak and spell the surname in a single utterance; and a Two Stage Speak and Spell strategy in which callers speak and spell the surname in two separate dialogue stages. The methodology employed in the research provides both quantitative user attitude data and performance results for each of the strategies, based on an empirical study with a cohort of 95 participants. The results show a clear distinction between strategies. User attitude towards the dialogues that involve both speaking and spelling the name is high. User attitude towards the Speak Only strategy is significantly less positive. Task completion rates are also significantly higher in the two strategies that involve spelling the name, at around 80% compared to just over 50% in the Speak Only strategy. The data underline the importance of user testing, demonstrating the value of the evaluation methodology used, and provide encouraging results for the strategies that involve both speaking and spelling the name.

show abstract

“…Automatic speech recognition of names from its spelling is an important sub-task for many applications such as directory assistance ( Lehtinen, 2000) and ( Schrâmm, 2000), or identification of city names for travel services (Lamel, 2000). Natural spelling implies the recognition of connected letters.…”

Section: Introductionmentioning

confidence: 99%