Handwritten Text Recognition (HTR)
AM Quartex enables transcription and full-text search of manuscript documents using Handwritten Text Recognition (HTR), an in-platform service customers can selectively apply. This proprietary neural network technology leverages artificial intelligence to recognize text in a multitude of scripts with remarkable accuracy, opening up manuscripts for detailed exploration by students and researchers.
Scroll down for more information about HTR, including samples - and be sure to check out the Manuscripts collection for more handwritten content.
Handwritten Text Recognition (HTR)
AM Quartex enables transcription and full-text search of manuscript documents using Handwritten Text Recognition (HTR), an in-platform service customers can selectively apply. This proprietary neural network technology leverages artificial intelligence to recognize text in a multitude of scripts with remarkable accuracy, opening up manuscripts for detailed exploration by students and researchers.
Scroll down for more information about HTR, including samples - and be sure to check out the Manuscripts collection for more handwritten content.
HTR Transcriptions
AM Quartex customers can use HTR to automatically generate transcriptions of English and foreign langage handwritten material. At the click of a button, HTR Transcription has the power to transform the close reading and research potential of manuscript collections. What's more, these transcripts can be downloaded by users or edited within Quartex to make alterations to enhance accuracy, meet screen-reader requirements and facilitate digital scholarship.
HTR Transcription also makes manuscript materials site searchable, optimising the discoverability of previously undiscovered or under-represented voices in your collections.
Some frequently asked questions about HTR can be read below.
HTR Transcriptions
AM Quartex customers can use HTR to automatically generate transcriptions of English and foreign langage handwritten material. At the click of a button, HTR Transcription has the power to transform the close reading and research potential of manuscript collections. What's more, these transcripts can be downloaded by users or edited within Quartex to make alterations to enhance accuracy, meet screen-reader requirements and facilitate digital scholarship.
HTR Transcription also makes manuscript materials site searchable, optimising the discoverability of previously undiscovered or under-represented voices in your collections.
Some frequently asked questions about HTR can be read below.
Whilst accuracy is, of course, dependant on legibility, orientation and condition of the script, HTR transcription boasts an impressive character accuracy rating of up to 97%. This innovative technology is also adept at recognising and replicating grammatical markers and character case.
How accurate is HTR?
Whilst accuracy is, of course, dependant on legibility, orientation and condition of the script, HTR transcription boasts an impressive character accuracy rating of up to 97%. This innovative technology is also adept at recognising and replicating grammatical markers and character case.
HTR transcripts can be generated for a wide range of languages using Western European character sets. HTR functionality identifies the language depicted on the page and replicates this in the transcript, incorporating diacritic glyphs and grammatical markers.
What languages can HTR Transcription read?
HTR transcripts can be generated for a wide range of languages using Western European character sets. HTR functionality identifies the language depicted on the page and replicates this in the transcript, incorporating diacritic glyphs and grammatical markers.
Yes! HTR Transcripts can be downloaded in a TXT format at the discretion of a Quartex website host.
Are HTR Transcripts eligible for download?
Yes! HTR Transcripts can be downloaded in a TXT format at the discretion of a Quartex website host.
Yes! The default HTR transcription style is proficient at recognising hands as early as the 15th Century through to the modern day. There are, however, two additional transcription styles available to users which are trained specifically across dense foreign language and black letter gothic script.
Does HTR Transcription work across manuscripts from any era?
Yes! The default HTR transcription style is proficient at recognising hands as early as the 15th Century through to the modern day. There are, however, two additional transcription styles available to users which are trained specifically across dense foreign language and black letter gothic script.
HTR Transcripts can be edited at any time through the Digital Asset Management System (DAMS) to enhance the transcript accuracy, by administrative or guest users.
For example, 'a Pity' in this manuscript has been transcribed as 'asity'. Adding in one space and correcting a single character would be very simple and improve the reader experience.
Can HTR Transcripts be edited?
HTR Transcripts can be edited at any time through the Digital Asset Management System (DAMS) to enhance the transcript accuracy, by administrative or guest users.
For example, 'a Pity' in this manuscript has been transcribed as 'asity'. Adding in one space and correcting a single character would be very simple and improve the reader experience.