• DOI:10.5772/6367
• Corpus ID: 3083954

Ultimate Trends in Integrated Systems to Enhance Automatic Speech Recognition Performance

@inproceedings{Durn2008UltimateTI,
  title={Ultimate Trends in Integrated Systems to Enhance Automatic Speech Recognition Performance},
  author={Carolina Palma Dur{\'a}n},
  year={2008}
}

• C. Durán
• Published 1 November 2008
• Computer Science

An automatic speech recognition (ASR) system can be defined as a mechanism capable of decoding the signal produced in the vocal and nasal tracts of a human speaker into the sequence of linguistic units contained in the message that the speaker wants to communicate (Peinado & Segura, 2006). The final goal of ASR is the man–machine communication. This natural way of interaction has found many applications because of the fast development of different hardware and software technologies. The most relevant are the access to information systems; an aid to the handicapped, automatic translation or oral system control. ASR technology has made enormous advances in the last 20 years, and now large vocabulary systems can be produced that have sufficient performance to be usefully employed in a variety of tasks (Benzeghiba et al., 2007; Coy & Barker, 2007; Wald, 2006; Leitch & Bain, 2000). However, the technology is surprisingly brittle and, in particular, does not exhibit the robustness to environmental noise that is characteristic of humans. Speech recognition applications that have emerged over the last few years include voice dialing (e.g., "Call home"), call routing (e.g., "I would like to make a collect call"), simple data entry (e.g., entering a credit card number), preparation of structured documents (e.g., a radiology report), domotic appliances control (e.g., "Turn on Lights" or "Turn off lights"), contentbased spoken audio search (e.g., find a podcast where particular words were spoken), isolated words with a pattern recognition, etc. With the advances in VLSI technology, and high performance compilers, it has become possible to incorporate different algorithms into hardware. In the last few years, various systems have been developed to serve a variety of applications. There are many solutions which offer small-sized, high performance systems; however, these suffer from low flexibility and longer cycle-designed times. A complete software-based solution is attractive for a desktop application, but fails to provide an embedded portable and integrated solution. Nowadays, High-end Digital Signal Processors (DSP’s) from companies, such as; Texas Instruments (TI) or Analog Devices and High-performance systems like Field Programmable Gate Array (FPGA) from companies, such as; Xilinx or Altera, that provide an ideal platform for developing and testing algorithms in hardware. The Digital signal processor (DSP) is one of the most popular embedded systems in which computational intensive algorithms can be applied. It provides good development flexibility O pe n A cc es s D at ab as e w w w .in te ch w eb .o rg