Are there any plans to automate some of the transcription with Apache Tika or tabula or some OCR tool?