International audienceEvaluating transcriptions from automatic speech recognition (ASR) systems is a difficult and stillopen problem, which often boils down to not considering only the word-error rate (WER). We present in this article a set of metrics, often used in other tasks in natural language processing (NLP), which we propose to apply in addition to WER in ASR. In particular, we introduce two measures relating to the morpho-syntactic and semantic aspects of transcribed words : 1) the POSER (Part-of-speech Error Rate), which highlights the grammatical aspects, and 2) the EmbER (Embedding Error Rate), an original measurement which takes up that of the WER by providing a weighting according to the semantic distance of the badly transcrib...
International audienceThe standard metric to evaluate automatic speech recognition (ASR) systems is ...
International audienceTexts generated by automatic speech recognition (ASR) systems have some specif...
With the development of technologies operating in a multilingual context, portability of speech tech...
International audienceEvaluating transcriptions from automatic speech recognition (ASR) systems is a...
International audienceEvaluating automatic speech recognition (ASR) systems is a classical but diffi...
International audienceEvaluating automatic speech recognition (ASR) systems is a classical but diffi...
International audienceEvaluating automatic speech recognition (ASR) systems is a classical but diffi...
International audienceThe standard metric used to evaluate the performance of Automatic Speech Recog...
International audienceThe standard metric used to evaluate the performance of Automatic Speech Recog...
International audienceThis paper investigates the evaluation of ASR in spoken language translation c...
Evaluation of automatic speech recognition (ASR) systems is difficult and costly, since it requires ...
We address the problem of estimating the quality of Automatic Speech Recognition (ASR) out-put at ut...
To transcribe speech, automatic speech recognition systems use statistical methods, particularly hid...
To transcribe speech, automatic speech recognition systems use statistical methods, particularly hid...
International audienceThe standard metric to evaluate automatic speech recognition (ASR) systems is ...
International audienceThe standard metric to evaluate automatic speech recognition (ASR) systems is ...
International audienceTexts generated by automatic speech recognition (ASR) systems have some specif...
With the development of technologies operating in a multilingual context, portability of speech tech...
International audienceEvaluating transcriptions from automatic speech recognition (ASR) systems is a...
International audienceEvaluating automatic speech recognition (ASR) systems is a classical but diffi...
International audienceEvaluating automatic speech recognition (ASR) systems is a classical but diffi...
International audienceEvaluating automatic speech recognition (ASR) systems is a classical but diffi...
International audienceThe standard metric used to evaluate the performance of Automatic Speech Recog...
International audienceThe standard metric used to evaluate the performance of Automatic Speech Recog...
International audienceThis paper investigates the evaluation of ASR in spoken language translation c...
Evaluation of automatic speech recognition (ASR) systems is difficult and costly, since it requires ...
We address the problem of estimating the quality of Automatic Speech Recognition (ASR) out-put at ut...
To transcribe speech, automatic speech recognition systems use statistical methods, particularly hid...
To transcribe speech, automatic speech recognition systems use statistical methods, particularly hid...
International audienceThe standard metric to evaluate automatic speech recognition (ASR) systems is ...
International audienceThe standard metric to evaluate automatic speech recognition (ASR) systems is ...
International audienceTexts generated by automatic speech recognition (ASR) systems have some specif...
With the development of technologies operating in a multilingual context, portability of speech tech...