Development of Speech Recognition Systems in Emergency Call Centers
نویسندگان
چکیده
In this paper, various methodologies of acoustic and language models, as well labeling methods for automatic speech recognition spoken dialogues in emergency call centers were investigated comparatively analyzed. Because the fact that dialogue has specific context noisy, emotional environments, available systems show poor performance. Therefore, order to accurately recognize speeches, main modules systems—language models training methodologies—as symmetric data approaches have been To find an effective model data, different types Gaussian Mixture Model/Hidden Markov Model (GMM/HMM) Deep Neural Network/Hidden (DNN/HMM) trained compared. Additionally, defined based on extrinsic intrinsic methods. Lastly, our suggested with spelling correction are compared common resulting outperforming other a notable percentage. Based results experiments, we determined DNN/HMM model, trigram Kneser–Ney discounting using before method configurations centers. It should be noted research was conducted two datasets collected from calls: Dialogue dataset (27 h), which encapsulates agents’ speech, Summary (53 contains voiced summaries those describing cases. Even though taken center is Azerbaijani language, belongs Turkic group languages, not tightly connected features. Hence, it anticipated can applied languages same group.
منابع مشابه
A framework for rapid development of conversational natural language call routing systems for call centers
A framework for rapid development of conversational natural language call routing systems is proposed. The framework cuts costs by using only scantily prepared business requirements to automatically create an initial prototype. Besides clear targets (terminal routing classes), vague targets which are variations of users’ incomplete (semantically overlapping) sentences are enumerated. The vague ...
متن کاملTelephone CPR Instructions in Emergency Dispatch Systems: Qualitative Survey of 911 Call Centers
INTRODUCTION Out-of-hospital cardiac arrest (OHCA) is a leading cause of death. The 2010 American Heart Association Emergency Cardiovascular Care (ECC) Guidelines recognize emergency dispatch as an integral component of emergency medical service response to OHCA and call for all dispatchers to be trained to provide telephone cardiopulmonary resuscitation (T-CPR) pre-arrival instructions. To beg...
متن کاملdevelopment of feminist poetics in adrienne rich
اشعار ریچ، به عنوان اشعاری که همیشه در حال تغییر و دگرگونی هستند، تجسمی از رشد و دگرگونیِ انسان هستد. پایان نامه ی حاضر، با تمرکز بر روی مراحل سیر شعری ریچ از تغییری در دنیا به عکس هایی فوری از یک عروس، سپس به شیرجه به درون کشتی شکسته و در نهایت به صبری عجیب مرا تا اینجا آورده، به بررسی این مراحل در قالب نظر شوالتر در رابطه با سه مرحله ی پیشرفت ادبی زنان یعنی مرحله ی زنانه، زن گرا و زن محور می پ...
15 صفحه اولEfficient language model development for spoken dialogue recognition and its evaluation on operator's speech at call centers
While a language model for recognition of spoken dialogue is ideally built from a very large, specific-task-oriented corpus, a great amount of time and effort is required to develop such a corpus, and this involves both the audio recording and written transcription of large amounts of speech data. Training data for a language model should match the target task in both topic and style. What is n...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Symmetry
سال: 2021
ISSN: ['0865-4824', '2226-1877']
DOI: https://doi.org/10.3390/sym13040634