Statement of Work
Project: AUTOMATED DTB PRODUCTION SOFTWARE


1.0.Background 
The NLS produces approximately 3,000 digital talking books (DTBs) annually. Traditional audiobook production requires substantial manual effort, including interaction with source materials to read and record narration and to create the XML control files that define the structure of each digital talking book. These processes are labor-intensive and limit the ability to scale production efficiently. 

To increase production capacity and efficiency, automation of the construction processes is required. XML control files of digital talking books shall be automatically generated from existing audiobooks and their associated metadata, as well as from electronic text source materials. In addition, narration shall be generated by leveraging synthetic speech systems to produce NLS announcements and/or the full-length audiobooks. 

The Library’s existing Digital Talking Book (DTB) production environment is a high-volume, semi-automated pipeline designed to transform source text into accessible, navigation-rich audio assets. The current DTB architecture utilizes a standardized ingestion-to-distribution framework that relies on adherence to established specifications, metadata schemas, and structural navigation hierarchies.

The primary components of this ecosystem include:
Content Pipeline: An interface for source content processing.
Storage and Data Management: An AWS S3-based architecture, used for both source content and final deliverables.
Production Workflow (manual/external): Application of the technical audio standards, including human narration, DTB mastering and loudness compliance (ITU-R BS.1770-5/EBU R128), and the embedding of internal metadata and integrity checksums.
Ingestion: Acquisition of the deliverable files that are fully compatible with current Library playback applications and distribution systems.

The objective of this solicitation is the integration of a new automated Text-to-Speech (TTS) DTB production capability that functions as a native extension of this existing workflow. The successful offeror must demonstrate technical compatibility with these existing production standards, specifically regarding text processing, metadata generation, and secure file packaging within the current operational environment

2.0.Scope
The contractor shall develop and integrate software to automate the production of accessible reading materials in accordance with all applicable standards and specifications. This effort includes the developing automated software to produce Digital Talking-Books (DTBs) composed of synthetic speech that complies with the National Library Service (NLS) ANSI/NISO Z39.86-2002 specifications. The solution shall enable efficient, scalable, and reliable production of these materials. The TTS software acquired under this contract shall be developed and delivered as a non-subscription software solution. The Library shall own the TTS software in its entirety, and the contractor shall not establish or convert the TTS software into a subscription-based product or service, and the contractor shall not retain any ownership interest in the TTS software.

3.0.Requirements
The contractor shall design software to automate the construction of digital talking books compliant with ANSI/NISO Z39.86-2002 and NLS specifications 1202, 1203, 1205, 1206 and D standards. The software shall be capable of generating digital talking books from commercial audiobooks with pre-existing audio files, as well as from electronic text source materials that do not include pre-existing audio. The software shall automate the synthesis of speech from provided announcement scripts and/or the electronic text source materials. 

3.1.Commercial Audiobooks with Supplemental Synthetic Speech Announcements
The contractor shall design and implement an automated tool to create digital talking books (DTBs) from commercial audiobooks with pre-existing audio files. At a minimum, the tool shall:

Access NLS repositories containing source audiobooks.
Extract catalog information and metadata associated with the source audiobooks.
Generate XML control files compliant with ANSI/NISO Z39.86-2002 and NLS specifications1201, 1202,1203. 
Populate OPF metadata elements using source metadata.
Identify navigable divisions within audio content and generate compliant navPoints, navTargets, and SMIL elements for DTB navigation.
Generate synthetic speech for required NLS announcements, navPoint audio elements, and any additional narration specified by the material.
Provide the capability to select synthetic voices for generated speech.
Support multiple audio file formats, including WAV, MP3, OPUS, and FLAC.
Convert source audio files to the AMR-WB+ file format, with an optional feature for MP3 encoding.
Apply encryption in accordance with PDTB2 and NLS 1205 specifications.
Validate all materials to ensure compliance with applicable standards and specifications.
Package completed materials for final delivery.
Generate production tracking reports and logs for auditing and quality assurance purposes.

The tool shall ensure efficient, reliable, and standards-compliant production of DTBs from commercial audiobook sources.

3.2. DTBs with Full-length Synthetic SpeechThe contractor shall design and implement an automated tool to create digital talking books (DTBs) from electronic text source material. At a minimum, the tool shall:

NLS will provide electronic text source materials.
Extract catalog information and metadata associated with the source audiobooks.
Generate XML control files compliant with ANSI/NISO Z39.86-2002 and NLS specifications.
Populate OPF metadata elements using source metadata.
Identify navigable divisions within the source material and generate compliant navPoints, navTargets, and SMIL elements for DTB navigation.
Provide the capability to generate various multimedia types in compliance with ANSI/NISO Z39.86-2002 standards.
Support text-based file formats, including but not limited to EPUB, Word Documents, Rich Text, and Plain Text.
Automate the generation of synthetic speech from electronic text source materials.
Provide functionality to select synthetic voices for narration.
Generate audio files in WAV format.
Generate audio files in AMR-WB+ format, with an optional feature for MP3 encoding.
Apply encryption in accordance with PDTB2 and NLS 1205 specifications.
Validate all materials to ensure compliance with applicable standards and specifications.
Package completed materials for final delivery.
Generate production tracking reports and logs for auditing and quality assurance purposes.

These capabilities shall ensure that the tool accommodates a wide range of text-based source materials and supports the creation of accessible, multimedia content aligned with the specified standards. 

3.3.Standards & Compliance Requirements
The contractor shall demonstrate full compliance with all applicable standards and shall be required to produce written certifications validating reports. Standards shall include but not limited to: 

NLS Specifications to include, 1202, 1203, 1205, 1206, current applicable versions, Contract Specifications – National Library Service for the Blind and Print Disabled (NLS) | Library of Congress
DAISY Standards the DAISY PDTB or successor formats, DAISY ANSI/NISO Z39.86-2002 – The DAISY Consortium
ANSI/NISO Z39.86-2002 accessibility standards, as applicable
Proper XML structure and validation for digital talking books

3.4. Functional Capabilities
The contractor shall provide a tool that can: 
Automatically generate digital talking books from commercial audiobooks with pre-existing audio files. 
Automatically generate and manage XML control/navigation files and Metadata and structural elements. 
Support synthetic speech generation for announcements and full-length digital talking books
Handle large-scale batch processing
Shall require demonstrations or samples. 

3.5.Automation & Scalability 
The contractor shall specify that the solution must: 
Minimize manual intervention
Support scalable, high-volume production
Operate reliably in production environments
Allow parallel processing where applicable
Support future growth without major redesign
Require performance benchmarks and throughput metrics. 

3.6.Security & Content Protection
The contractor shall protect accessible materials which require: 
Secure handling of source files and outputs
Encryption 
Compliance with applicable federal IT security requirements. See Section H for the applicable security contract requirements. 
Clear data retention and destruction procedures

3.7. Interoperability & Integration
The contractor shall demonstrate that the software: 
Integrate with NLS AWS systems.
Supports standard file formats (audio, XML, text, metadata)
Allows export/import without proprietary lock-in
Requires interface controls. 
The contractor shall demonstrate knowledge and experience with Amazon Web Services (AWS) sufficient to support the requirements of this contract.

3.8. Documentation, Training & Support 
The contractor shall: 
Deliver user manuals and technical documentation
Provide administrator and operator training
Provide ongoing technical support and maintenance
Update and patch management process
Contractor shall provide technical support via email, portal and/or telephone. 
Contractor’s technical support response times shall be within 24 hours of support request. 

3.9. Warranty
The contractor shall warrant: 
All software, tools, and services provided under this contract shall conform to the requirements of the SOW and shall be free from defects in materials, workmanship, for a period of twelve (12) months follow Government acceptance of supplies and services. 
Further warrants that the software will comply with all applicable accessibility and production standards. 
During warranty period, the contractor shall at no additional cost to the Government promptly correct any defects, nonconformances, or performance failures and provide necessary updates or patches. 
This warranty includes non-infringement of third-party intellectual property. 

3.10. Status Reports

Status reports shall include, but should not be limited to:
Project Progress: A summary of the project’s current status, including milestones, deliverables, and any deviations from the plan.
Quality Control: An overview of identified risks, their impact on the project, and mitigation strategies.
Action Items: A list of tasks and milestones that need to be addressed to keep the project on track.

3.11.Quality Control Report/Procedures
The contractor shall implement Quality Control Procedures (QCP) in a TTS-to-DTB production system to ensure generated DTBs meet audio quality, structural, accessibility, and technical standards before delivery, and any other quality control procedures necessary. 
The contractor shall perform audio quality checks to ensure synthesized speech is clear, intelligible, and free from distortion of processing errors. 
The contractor shall verify all DTB files according to the required DTB structure and navigation formats. 
The contractor shall verify that all metadata associated with each DTB file is accurate and correctly formatted. 
The contractor shall confirm all files are complete and free of corruption. 
The contractor shall verify that the generated DTB files function correctly with DTB playback systems and applications used by the Library.
The contractor shall identify and correct errors discovered during the quality control process before final delivery. 
3.12.Software Ownership and Non-Subscription Requirement
The software acquired under this contract shall be developed and delivered as a non-subscription software solution. 
The contractor shall not provide, establish, or convert the TTS software into a subscription-based service, Software-as-a-service (SaaS) offering, or other recurring license-based product or service. 
The Library is purchasing the TTS software and shall own the resulting software in accordance with the contract. 
The contractor shall not require the Library to purchase, renew, or maintain any separate software licenses, subscriptions, or other recurring access rights to use the software. 
Ongoing operational support, maintenance, and technical support services provided under this contract are separate services and shall not constitute a subscription, license, or recurring software access fee. 
No follow-on purchase, renewal, subscription, or license requirement shall be created as a condition of the Library’s continued use or operation of the TTS software. 
The contractor shall not retain or assert any ownership interest in the TTS software. 


PHASES I, II, III, AND IV – NON-RECCURING TTS SOFTWARE DESIGN AND DEVELOPMENT, SYSTEM INTEGRATION & OPERATIONAL TESTING AND FULL SYSTEM INTEGRATION & DEPLOYMENT AND ONGOING OPERATIONAL SUPPORT. 

3.13.Phase I: Non-Recurring TTS Software Design and Development
The non-recurring development phase includes all one-time activities required to software design and development the Text-to-Speech (TTS) capability for the Digital Talking Books (DTBs). 

3.13.1.System Design and Technical Approach
The Contractor shall provide a System Design and Technical Approach in non-recurring development phase of the new TTS system for DTB the contractor shall explain how the system will work, how it will integrate with DTB and how it will meet the requirements. 

Architecture Diagrams: The contractor shall provide architectural diagrams for the system to provide clear visual representation of how the TTS system will be structured and how it will integrate with the Library’s storage and distribution. Architecture documentation shall include, (1) a System Content Diagram, (2) a Component Architecture Diagram, (3) a Data Flow Diagram, (4) a DTB Packaging and Output Structure Diagram, and (5) a Security and Access Architecture Diagram.  
System Design Technical Documentation: 
 System Architecture Description Document. The Contractor shall provide a system architecture description document identifying the major components required to implement the Text-to-Speech (TTS) capability and how those components interact with the Digital Talking Book (DTB) system, including architecture, components, data flow, integration, interfaces, and environment for the TTS-DTB system. This document explains the overall technical structure of the system. It focuses on how the TTS system is built. 
DTB Integration Approach Document The description shall address text ingestion, speech synthesis, generation of DTB navigation elements and metadata, packaging of DTB-compatible files, and integration with existing Library storage and distribution and playback applications. The contractor shall also identify any system interfaces or workflows required to support Library’s production control, ingest, storage and distribution using the TTS capability. 
Text Processing Workflow Document: The contractor shall describe the processing workflow used to convert text-based content into synthesized speech, including text ingestion, language processing, voice generation, and production of DTB-compatible audio output. The contractor shall explain how source text will be prepared, processed, and converted into speech before it becomes a Digital Talking Book (DTB) to ensure the text is handled correctly so generated audio is accurate, structured, and compatible with DTB standards. The contractor shall describe the text processing workflow used to convert source text into synthesized speech. The description shall address text ingestion, preprocessing and normalization, segmentation of content, speech synthesis, audio post-processing, and quality control measures necessary to produce DTB-compatible audio output. This document ensures the Library understands exactly how text becomes a Digital Talking Book before development proceeds.  
Audio Output Specifications Document: The contractor shall identify the audio output formats and quality specifications used to generate synthesized speech compatible with Digital Talking Book playback systems. The contractor shall identify the audio format used for DTB production (e.g., WAV, MP#, or other DTB-compatible formats). The contractor shall specify audio encoding (e.g., Bitrate, sampling Rate, Channel configuration, compression method). The contractor shall describe how audio will be divided into sections such as chapters, headings, or navigation points to support DTB navigation. The contractor shall describe any processing applied to the generated speech, including volume normalization, noise reduction, and audio leveling. The contractor shall identify available voice options and ensure the synthesized speech is clear, intelligible, and appropriate for accessible talking book playback. The contractor shall ensure that the generated audio files are compatible with DTB playback applications and devices used by the Library. 
Metadata and File Packaging Description The contractor shall define how metadata, navigation elements, and structural information will be generated and incorporated into Digital Talking Book files to ensure proper playback and navigation. The contractor shall describe the Metadata Generation and how the system will create and manage metadata associated with each DTB, including titles, authors, identifiers, and other descriptive information required for cataloging and playback. The contractor shall describe the Navigation Elements and how the system will generate structural navigation elements such as chapters, headings, page markers, and other navigation points used within DTB files. The contractor shall describe file naming conventions used to ensure consistency and compatibility with the Library’s DTB production and storage environment. The contractor shall describe the storage and output organization how completed DTB packages will be organized and prepared for ingestion into the Library’s cataloging or distribution systems.
System Security and Data Protection Description Document. The contractor shall describe the security measures implemented to protect source content and generated audio files during processing, storage, and system operation. This ensures that copyrighted materials, system data, and user access are properly controlled and safeguarded. The contractor shall describe access control and how access to TTS system will be restricted to authorized personnel through authentication and role-based permissions. The contractor shall describe data protection and how source text files, generated audio files, and DTB outputs will be protected during processing, storage, and transfer. The contractor shall describe secure data transmission for how data will be protected when transferred between system components or external systems, including encryption methods if applicable. The contractor shall describe system logging and monitoring and how system activity will be logged and monitored to detect unauthorized access or system issues. The contractor shall describe content protection and how copyrighted source materials and generated DTB files will be safeguarded from unauthorized access or distribution. The contractor shall describe backup and recovery and how the contractor will protect against data loss through backup procedures and system recovery capabilities. 
Testing and Validation Plan Document: The contractor shall provide a testing and validation approach demonstrating how the Text-to-Speech system will be verified to meet functional, accessibility, and Digital Talking Book compatibility requirements. How with the contractor verify the Text-to-Speech (TTS) system functions correctly and produces Digital Talking Book (DTB) outputs that meet the Library’s technical and accessibility requirements. The contractor shall provide a functional testing plan to describe how the TTS system will be tested to confirm that all features operate as intended. The contractor shall explain how generated DTB files will be tested to ensure compatibility with DTB playback applications and navigation requirements. The contractor shall provide audio quality validation by describing how the quality and intelligibility of synthesized speech will be evaluated to ensure clear and accessible audio output. The contractor shall provide metadata and navigation verification to verify that DTB navigation elements (chapters, sections, headings) and metadata are correctly generated and structured. The contractor shall provide error handling and defect resolution to identify procedures for detecting system errors, documenting defect and correcting issues identified during testing. 
Implementation Schedule Plan Document: The contractor shall provide a project schedule identifying development milestones, system integration activities, and testing timelines for delivery of the operational Text-to-Speech capability. 
TTS Prototyping, Development, Testing and Samples 
TTS Prototype: The contractor shall develop an initial working prototype of the TTS capability. The purpose is to demonstrate text-to-speech generation, confirm voice quality and intelligibility and validate basic processing workflow. The contractor shall produce a prototype demonstrating the integration of Text-to-Speech generated audio into the DTB structure, including navigation elements and metadata. 
DTB Prototype Output: The contractor shall produce sample DTB for evaluation of audio quality, navigation structure, and DTB compatibility. 
Sample Speech Outputs: The contractor shall provide synthesized speech samples using available voice options to support evaluation of clarity, intelligibility, and suitability for accessible talking book production. 
Workflow Demonstration: Samples shall demonstrate the end-to-end workflow used to convert source text into synthesized speech and package the output as a DTB. 
Testing Results Report: The contractor shall provide testing results documenting system functionality, DTB file compatibility, and audio output validation. 

Phase II: System Integration and Operational Testing 
Phase 2 shall focus on system integration and operational validation of the Text-to-Speech capability within the DTB production environment prior to full production use. Phase 2 allows the Library to verify the system works with existing infrastructure, ensures DTB files function correctly in playback systems, and identify and fix integration issues before large-scale use. 
System Integration Report with Library’s storage and distribution: The contractor shall integrate the TTS system into the Library’s DTB production environment and ensure compatibility with existing systems, workflows, and DTB playback applications.  
Interface and Workflow Integration: The contractor shall configure system interfaces and automated workflows required to support text ingestion, speech synthesis, DTB file generation, and transfer of completed DTB packages to the Library’s storage or distribution systems. 
Operational Testing Results: The contractor shall perform operational testing using representative content to verify that the TTS system produces valid DTB outputs and operates correctly within the Library’s storage and distribution. 
Sample DTB Outputs: The contractor shall validate that DTB outputs generated by the system comply with DTB file structure requirements and function correctly with DTB playback systems used by the Library. 
Performance and Reliability Testing: The contractor shall evaluate system performance and reliability under operational conditions to ensure the system can support routine DTB production. 
Defect Resolution and System: The contractor shall document and resolve defects identified during system integration and operational testing to ensure proper system functionality. 
Integration Completion Report: The contractor shall document the completion of the integration. 

Phase III: Full System Integration and Full Deployment
For Phase 3-Full Integration and Full Deployment, the contractor shall operate the fully integrated TTS system to produce DTB at scale. Phase 3 moves the system from testing into full operational use. 
Full Integration & Full Deployment Operations: The contractor shall proceed with full integration and deployment operations for the TTS system to generate DTB files according to the Library’s requirements. 
DTB Production Workflow Management: The contractor shall manage the operational workflow used to convert source text into synthesized speech and package the results into DTB-compatible files for delivery to the Library’s storage and distribution systems. 
System Monitoring and Operational Support: The contractor shall monitor system performance and provide operational support necessary to maintain reliable full integration and full deployment of DTBs. 
Quality Control and Validation: The contractor shall perform quality control checks to ensure generated DTB files meet the Library’s technical and accessibility standards. 
Phase IV: Ongoing Operational Support
For Phase 4-Ongoing Operational Support, the contractor shall provide ongoing operational support which includes but not limited to, operations, maintenance, technical support, security, quality control, system monitoring, to maintain the TTS software as a reliable and fully operational component of the Digital Talking Book production environment throughout the remainder of the period of performance. Phase 4 moves the system into ongoing operational support. 

The TTS for Digital Talking Book System, ongoing operational support shall include but not limited to: 

System Operations and Maintenance
System monitoring and performance management
Routine maintenance and software updates
Bug and defect resolution

TTS and DTB Production Support
TTS engine configuration and tuning 
DTB production support
Audio quality monitoring
Metadata, navigation, and file validation
Compliance with applicable DTB standards

System Integration and Technical Support
API and interface maintenance
Integration troubleshooting
Technical support and issue resolution
Compatibility testing
Release and deployment support

Security and Data Protection
Security monitoring
Vulnerability remediation
Access control
Data and content protection
Backup and disaster recovery

Quality Assurance and Performance Management
Quality control procedures
Production and system performance monitoring
Incident tracking and corrective actions
Capacity and scalability monitoring

Configuration Management
System configuration changes
Updates required to maintain interoperability
Performance improvements

Documentation, Training, and Reporting
Technical documentation updates
User training and refresher training
Status and performance reports
Maintenance and change records

4.0.Contract Performance
4.1.Schedule of Events/Deliverables 

REFERENCE
DELIVERABLES
DUE DATE
3.9.
12-Month Warranty
Warranty period shall commence upon written Government acceptance of all deliverables and shall remain in effect for twelve (12) months thereafter.
3.10.
Status Reports
Submit status report every thirty (30) calendar days throughout the period of performance.
3.11.
Quality Control Reports
Included with Submission of Status Reports every thirty (30) calendar days throughout the period of performance.
3.13.1.
Phase I: Architectural Diagrams
Due within 30 calendar days after the start of the period of performance.
3.13.1.
Phase I: Documentation: (1) System Architecture, (2) DTB Integration, (3) Text Processing Workflow, (4) Audio Output Specification, (5) Metadata and File Packaging, (6) System Security and Data Protection, (7) Testing/Validation Plan, (8) Schedule Plan
Due within 30 calendar days after the start of the period of performance.
3.13.2.
Phase I: TTS Prototype, DTB Prototype Output and Sample Sets, Workflow Demonstrations, Testing Results Report
Due within six (6) months after the start of the period of performance. 
3.14.
Phase II: System Integration Report, Operational Testing, Sample Outputs, Integration Completion Report
Phase II shall begin immediately upon successful completion and Government acceptance of Phase I and shall be completed within ninety (90) calendar days
3.15.
Phase III: Full Integration and Full Deployment, DTB Integration/Deployment Workflow, Quality Control/Validation, Documentation Updates
Phase III shall begin immediately upon successful completion and Government acceptance of Phase II and shall be completed within ninety (90) calendar days. 
3.16.
Phase IV: Ongoing Operational Support
Phase IV shall begin immediately upon successful completion and Government acceptance of Phase III and shall continue for the remainder of the period of performance. 
5.2
Kick Off Meeting
Due within ten (10) business days after start of POP date
6.1
Key Personnel Resumes
Due with Proposal Submission

5.0.Additional Information 
5.1. Government Furnished Property (GFP) 
NLS may make digital source material or samples available to aid in tasks required in this contract. All digital source material is the property of the Library of Congress. The contractor shall ensure that the digital source material or samples are not used for any purpose other than contract fulfilment. The contractor shall securely dispose of the digital source material at the end of the contract.

5.2. Kick-Off Meeting (Post-Award) 
A Kickoff meeting shall be held within ten (10) business days of contract award. The kick-off meeting shall be coordinated with the designated by the COR. The kickoff meeting shall be conducted virtually via Teams platform. Attendance by the contract administrator is required. Kickoff meeting will provide an opportunity for both the contractor and Library to discuss logistics, address questions, and clarify any contract requirements.

6.0.Special Qualifications 

6.1.Key Personnel Resumes
The following position(s) are considered key personnel: 
Project Manager (PM): 
The contractor shall designate a Project Manager (PM) who will serve as the primary point of contact for discussions with the Contracting Officer and the COR. The PM shall be responsible for overall communications, timeliness, and overall quality assurance of all production activities.  Oversee the management software development or IT system integration. Experienced coordinating development, testing, and deployment activities. Experience with accessibility and/or digital media systems. 
Qualifications: 
Education: Bachelor’s Degree in relevant field
Experience: The Project Manager shall have at least 3 to 5 years or relevant experience managing software development and/or system integration projects and shall be responsible for coordinating project activities, schedules and deliverables. 
Senior Engineer: 
The Engineer shall have experience developing or implementing speech synthesis technologies and configuring systems used to generate synthesized speech. They shall have experience integrating software systems and developing automated workflows for digital content processes and have experience evaluating and processing digital audio to ensure clarity, consistency, and compliance with required audio standards. 
Qualifications: 
Education: Bachelor’s Degree in relevant field
Experience: 3 to 5 years’ experience in speech technology and/or TTS development, software development, and audio engineering.

[End of Statement of Work]