source encoding for spanish

Hello, The content is showing strange symbols like "formación desde hace 25 años". The web where I extract info is in spanish and it has <meta <span class="html-attribute-name">charset</span>="<span class="html-attribute-value">utf-8</span>"> And in the <b>Source Article Encoding</b> option I tried:

• Auto detect
• UTF-8
• ISO-8859-1
• ...

But I the content extracted has always some strange symbols. Any idea about it?