Can you provide the text in the FAX headers as output from spandsp when receiving a FAX?

When a FAX is received, you might see a header line on each page. These headers are received graphically, as part of the page images. As such, they end up in the received TIFF file, as part of the page images. The text of these headers is not available as ASCII/Unicode text as the FAX is received. The only way to make it available as text would be to implement an optical character recognition (OCR) scheme. For various reasons, of complexity speed and accuracy, doing this inside the FAX software is not good idea, and is not implemented.

If OCR is applied to the TIFF file resulting from a received FAX, reliably extracting useful information from the header line would be difficult. The header may or may not be present (although it is a legal requirement to have some kind of header in some countries). The format of the strings, the font, and the size of the header are not standardised. The header might not even contain Romance text.

The only information sent between the FAX machines as text is an optional 20 character identifier. The T.30 standard says this should contain digits, and it is usually set to the telephone number of the FAX machine. In fact, most machine can handle any ASCII characters in this identifier. This text is the information typically displayed on the small LCD screen of a FAX machine. Spandsp makes the received identifier strings available to the application.