Loading...
Please wait, while we are loading the content...
Similar Documents
Improvement of Speech Intelligibility of Mobile Devices in Noisy Environments
| Content Provider | Semantic Scholar |
|---|---|
| Copyright Year | 2016 |
| Abstract | Multi-channel signal processing has gained popularity due to its practical application (e.g., speech recognition, noise reduction etc.) in a wide range of fields (i.e., military applications, music editing, audio information retrieval, etc.), and to the increased use of mobile and assistive communication devices. Without a doubt, increases in the computational power of these devices have allowed for the use of a wide range of techniques for noise reduction and to increase speech intelligibility. However, many of these are ineffective in many situations and, if not adjusted correctly for the particular environment, often introduce artifacts and distortions to the signal. Furthermore, these devices are often expensive, removing there availability from the general public, and proprietary, making them difficult to replicate and verify. Hence, there is a need for techniques for improving speech intelligibility in noisy environments which are open source, verifiable, and repeatable. Objectives and Importance The overall objective is to create a technique for improving speech intelligibility in noisy environments. To achieve this, the following subtopics must be addressed. Sound Source Separation Source separation problems in Digital Signal Processing (DSP) are those in which several signals have been mixed together into a combined signal and the objective is to recover the original component signals from the combined signal [1], [2]. The classical example of a source separation problem is the cocktail party problem, where a number of people are talking simultaneously in a room (e.g., at a cocktail party), and a listener is trying to follow one of the discussions. The human brain can handle this sort of auditory source separation problem, but it is a difficult problem in DSP. Several approaches have been proposed for the solution of this problem but development is currently still very much in progress. Some of the more successful approaches are Principal Component Analysis (PCA) and Independent Components Analysis (ICA), which work well when there are no delays or echoes present (i.e., the problem is simplified a great deal). The field of computational auditory scene analysis attempts to achieve auditory source separation using an approach that is based on human hearing. The human brain must also solve this problem in real time. In human perception, this ability is commonly referred to as Auditory Scene Analysis (ASA) or the cocktail party effect. These techniques will be investigated for separating an audio mixture and then HeadRelated Transfer Functions (HRTFs) will be used to position the sources in their respective positions in 3D space. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://international.fullerton.edu/pdf/siri/project12.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
