Towards a speech-to-speech machine translation quality metric</tile> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> <meta http-equiv="description" content="Towards a speech-to-speech machine translation quality metric"> <meta http-equiv="keywords" content="Towards,a,speech-to-speech,machine,translation,quality,metric"> </head> <body> <table width="100%" border="0" cellspacing="0" cellpadding="0"> <tr> <td width="10" height="39"> </td> <td width="99%" height="39"> <table width="100%" border="0" cellspacing="0" cellpadding="0"> <tr> <td width="16%"><a href="http://inttranews.inttra.net"><img src="http://inttranews.inttra.net/images/inttranews_grey_2.jpg" width="192" height="37" border="0"></a></td> <td width="50%" valign="middle"> <p><font face="Arial, Helvetica, sans-serif" size="2"><b><font color="#CCCCCC" face="Verdana, Arial, Helvetica, sans-serif" size="1">KNOW.<br> HOW.</font></b></font> </p> </td> <td width="34%" align="right" valign="top"><font face="Verdana, Arial, Helvetica, sans-serif" size="1" color="#666666">27-06-2005</font></td> </tr> </table> </td> </tr> <tr> <td width="10"> </td> <td width="99%" > </td> </tr> <tr> <td width="10"> </td> <td width="99%" > <table width="100%" border="0" cellspacing="0" cellpadding="0"> <tr> <td width="100%"><font face="Verdana, Arial, Helvetica, sans-serif" size="2"> <b>Towards a speech-to-speech machine translation quality metric</b></font> </td> </tr> <tr> <td width="100%"> </td> </tr> <tr> <td width="100%"> <font face="Verdana, Arial, Helvetica, sans-serif" size="2">New Jersey, USA (Lockheed Martin Advanced Technology Laboratories): In this research paper, the general characteristics of a pragmatic metric for the production evaluation of speech-to-speech translations are discussed. While these characteristics constrain the space of allowable metrics, infinite definition space remains from which to select and define any particular metric. The recommended characteristics are drawn from the author’s experience as primary developer of a text-based translation quality metric used in a production environment. The primary contribution is that of strict category ordering and two meta-rules that reduce the variance in assignment of errors to categories. <br> <br>For more information, please visit: <br><a href="http://www.atl.external.lmco.com/overview/papers/1116.pdf " target="blank">www.atl.external.lmco.com/overview/papers/1116.pdf </a> <br></font></td> </tr> </table> </td></tr> </table> <script type="text/javascript"></script> <script type="text/javascript" src="<http://pagead2.googlesyndication.com/pagead/show