Today's Challenges for Embedded ASR
نویسندگان
چکیده
Automatic Speech Recognition (ASR) is pervading nowadays to areas unimaginable a few years ago. This progress was achieved mainly due to massive changes in the “Smart phones world” and ubiquitous availability of small, and powerful Linux-based hardware. Recently, the availability of free ASR systems with acceptable speed and accuracy performance grew. Together with the changes the mobile world brought, a developer is now able to include ASR quickly without detailed knowledge of the underlying technology. What will be the future of embedded ASR systems in this case? This talk presents two embedded ASR applications and points out their advantages over today’s quick solutions. The first one demonstrates how changes in users behavior allowed to design a usable voice enabled house control. The second one is an extremely reliable in–car real–time ASR system which can even use a remote ASR for complex tasks.
منابع مشابه
Towards large vocabulary ASR on embedded platforms
In this paper we present an overview of an automatic speech recognition system implementation in the context of embedded systems. Specific challenges presented by low resource platforms will be addressed for the basic components of an ASR decoder. Our main objective is to utilize and modify the technology developed for large vocabulary ASR to achieve efficient LVCSR on embedded systems as well.
متن کاملA Novel Speech/Noise Discrimination Method for Embedded ASR System
The problem of speech/noise discrimination has become increasingly important as the automatic speech recognition (ASR) system is applied in the real world. Robustness and simplicity are two challenges to the speech/noise discrimination method for an embedded system. The energy-based feature is the most suitable and applicable feature for speech/noise discrimination for embedded ASR system becau...
متن کاملSpoken Language Translation for Polish
Spoken language translation (SLT) is becoming more important in the increasingly globalized world, both from a social and economic point of view. It is one of the major challenges for automatic speech recognition (ASR) and machine translation (MT), driving intense research activities in these areas. While past research in SLT, due to technology limitations, dealt mostly with speech recorded und...
متن کاملAmbient Intelligence with the Ubiquitous Network, the Embedded Computer Devices and the Hidden Databases (abstract)
Within 10 years people will have hundreds of interconnected computing devices around them. Many connections will be wireless and many devices will store data in them. With a billion or more mobile users having a hundred devices we face a ubiquitous (IP) network with a very large (100B) small and hidden or embedded databases as well as \always-on" connection to all kind of databases in the netwo...
متن کاملFpga-based Embedded Solution for Automatic Speech Recognition
Speech recognition can prove a valuable asset to many industrial processes as it is a natural form of command input. Automatic speech recognition (ASR) is the ability of machines to derive commands from input speech. ASR systems can ease the amount of training required by system operators, as one simply needs to say what needs to be done. A typical speech recognition system can be segmented int...
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 2014