Description
The Library of Congress (LOC) is conducting market research for an automated Text‑to‑Speech (TTS) system to produce Digital Talking Books (DTBs) that meet DAISY ANSI/NISO Z39.86‑2002 and NLS specifications. The contractor would design, develop, and integrate non‑subscription software that can (1) generate DTBs from commercial audiobooks with existing audio, (2) create DTBs from electronic text using synthetic speech, (3) automatically produce XML control/navigation files, metadata, and encryption, and (4) support large‑scale batch processing on the LOC’s AWS‑based infrastructure. The effort is broken into four phases: (i) non‑recurring software design and prototype, (ii) system integration and operational testing, (iii) full deployment, and (iv) ongoing operational support and maintenance, including a 12‑month warranty. The LOC seeks capability statements, rough order‑of‑magnitude pricing, and feedback on the draft Statement of Work. Participation is voluntary and does not guarantee a future award.
• Scope: Design & deliver a stand‑alone TTS application owned outright by the LOC (no SaaS/subscription).
• Key functionalities: audio format conversion, synthetic voice selection, XML/SMIL generation, encryption, batch processing, AWS integration.
• Deliverables: architectural diagrams, design documents, prototypes, testing/validation reports, status reports (every 30 days), and a 12‑month warranty.
• Target volume: ~3,000 DTBs per year, requiring high‑throughput automation.
• Required expertise: custom software development, speech synthesis, DAISY/DTB standards, AWS cloud services, security/compliance.
Attached files
Contact
James Robinson