N-best: the northern- and southern-dutch benchmark evaluation of speech recognition technology

Kessens, Judith M.; Leeuwen, David A. van

doi:10.21437/interspeech.2007-413

Cited by 12 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first set of experiments are performed using our broadcast news development set. The second set of experiments is conducted on the recently made available development set for the Dutch benchmark N-Best [9].…”

Section: Methodsmentioning

confidence: 99%

Fast n-gram language model look-ahead for decoders with static pronunciation prefix trees

Huijbregts¹,

Ordelman²,

Jong³

2008

Interspeech 2008

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 99%

Fast n-gram language model look-ahead for decoders with static pronunciation prefix trees

Huijbregts¹,

Ordelman²,

Jong³

2008

Interspeech 2008

View full text Add to dashboard Cite

Introduction

Spyns

2012

Essential Speech and Language Technology for Dutch

View full text Add to dashboard Cite

The STEVIN ("STEVIN" is a Dutch acronym for "Essential Speech and Language Technology Resources for Dutch") programme aimed to contribute to the further progress of Human Language Technology for Dutch (HLTD) in the Low Countries (i.e., Flanders and the Netherlands) and to stimulate innovation in this sector. The major scientific goals were to set up an effective digital language infrastructure for Dutch, and to carry out strategic research in the field of language and speech technology for Dutch. 1 Consortia could submit project proposals in response to calls for proposals. Several calls were issued, and they included three open calls and two calls for tender as well. The thematic priorities for each call were determined in line with the overall STEVIN priorities and the state of their realisation before each call. The STEVIN thematic priorities, based on what is called the Basic Language Resource Kit (BLARK) for Dutch [20], are summarised in Tables 1.1 and 1.2. A BLARK is defined as the set of basic HLT resources that should be available for both academia and industry [13].STEVIN advocated an integrated approach: develop text and speech resources and tools, stimulate innovative strategic and application-oriented research, promote embedding of HLT in existing applications and services, stimulate HLT demand via 1 We refer the reader to Chap. 2 for more details. P. Spyns ( ) Nederlandse Taalunie,

show abstract

N-Best 2008: A Benchmark Evaluation for Large Vocabulary Speech Recognition in Dutch

Leeuwen¹

2012

Theory and Applications of Natural Language Processing

View full text Add to dashboard Cite

Automatic Speech Recognition (ASR) is a discipline of engineering that benefits particularly well from formal evaluations. There are several reasons for this. Firstly, speech recognition is basically a pattern recognition task, and to scientifically show that the system works it needs to be tested on fresh material that has never been observed by the system, or indeed the researchers themselves. This means that speech material for testing purposes needs to be collected, which requires quite some effort, but can formally only be used once. It is therefore more efficient if the evaluation material is used to determine the performance of several systems simultaneously, which suggests a common form of this kind of performance benchmarking: that of a formal evaluation. Secondly, after a system evaluation the evaluation material and protocol can be used for future researchers as a benchmark test: algorithms can be developed and tuned to increase performance on the test. By using a well-established formal evaluation protocol performance figures can directly be compared amongst different researchers in the literature, which gives more meaning to the actual figures. Thirdly, a benchmark evaluation gives researchers a clear focus and goal, and appears to stimulate the different research groups to get the best out of their system in a friendly competitive way.Formal evaluations in speech technology have their origin in the early 1990s of the last century, when the US Advanced Research Projects Agency (ARPA) organised regular evaluations in speech recognition executed by the National Institute of Standards and Technology (NIST) [16], soon followed by speaker [12] and language [13] recognition. In the early years the language of interest for speech recognition invariably was English, but as tasks got harder and performance got

show abstract

N-best: the northern- and southern-dutch benchmark evaluation of speech recognition technology

Cited by 12 publications

References 13 publications

Fast n-gram language model look-ahead for decoders with static pronunciation prefix trees

Fast n-gram language model look-ahead for decoders with static pronunciation prefix trees

Introduction

N-Best 2008: A Benchmark Evaluation for Large Vocabulary Speech Recognition in Dutch

Contact Info

Product

Resources

About