Sage: The New BBN Speech Processing Platform
نویسندگان
چکیده
To capitalize on the rapid development of Speech-to-Text (STT) technologies and the proliferation of open source machine learning toolkits, BBN has developed Sage, a new speech processing platform that integrates technologies from multiple sources, each of which has particular strengths. In this paper, we describe the design of Sage, which allows the easy interchange of STT components from different sources. We also describe our approach for fast prototyping with new machine learning toolkits, and a framework for sharing STT components across different applications. Finally, we report Sage’s state-of-the-art performance on different STT tasks.
منابع مشابه
Applications of the BBN Sage Speech Processing Platform
As a follow-up to our paper at Interspeech 2016 [1], we propose to showcase various applications that now all use BBN’s Sage Speech Processing Platform, demonstrating the platform’s versatility and ease of integration. In particular, we will showcase 1) BBN TransTalk: A turnbased speech-to-speech translation program running entirely on an Android smartphone, alongside a custom 3D-printed periph...
متن کاملWest Point, SAAVB, and BBN/AUB Arabic Speech Corpora: A Comparative Survey
The aim of this paper is to evaluate three public Arabic speech corpora, namely the West Point (WP), Saudi Accented Arabic Voice Bank (SAAVB) and the BBN Technologies/American University at Beirut (BBN/AUB) corpus by referring the TIMIT English speech corpus as benchmark. Weaknesses, strengths, and discrepancies of these Arabic corpora regarding their design and content are covered in this pape...
متن کاملPorting tonNew Domains Using the Learner
Acquiring syntactic and semantic information about a new application domain for a natural language processing system is often a time-consuming task. To address this problem, various researchers have developed acquisition tools to speed the process. While such tools are very useful, they are typically tied to particular systems and so their benefits cannot be shared by other researchers. In this...
متن کاملAuditory Cortical Temporal Processing Abilities in Young Adults
Purpose: To evaluate whether cortical encoding of temporal processing ability, using the N1 peak of the cortical auditory evoked potential, could be measured in normally hearing young adults using three paradigms: voice-onsettime, speech-in-noise and amplitude-modulated broadband noise. Research design: Cortical auditory evoked potentials (CAEPs) were elicited using: (1) naturally produced stop...
متن کاملA Speech Processing Research Platform for Android Based Smart Phones and Tablets
This paper presents a new research and education platform for speech processing. The platform is called Speech Enhancement for Android (SEA) and incorporates past and present speech enhancement techniques applied to recorded speech corrupted by real world noise sources. Researchers, students and teaching staff can use this platform to perform speech enhancement and observe its effects on the li...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2016