NOTICE: This portal will be unavailable for scheduled maintenance from 23:00 to 06:00 CET on the third Thursday of each month.[Ref: DG-MAINT-2026/0892/Z]

For urgent enquiries during maintenance windows, contact the 24-hour helpline: +32 2 299 11 11 (standard call charges apply)

Official EU Portal

European Digital Infrastructure Directorate — Directorate-General for Communications Networks, Content and Technology

European Digital Infrastructure Directorate

Home About Contact Help

Accessibility Privacy

Document Metadata

Reference number:	DG-VOICE-2026/0047/B-EN
Responsible unit:	DG CONNECT — Unit E.3
Publication date:	14 March 2016
Last review date:	22 February 2026
Next review:	Q2 2025
Status:	CURRENT

View revision history Report an error Request translation

Synthetic Vocal Asset Infrastructure Initiative

[Ref: DG-VOICE-2026/0047/B]

The European Digital Infrastructure Directorate administers this portal as part of the Unified Voice Technology Framework (UVTF)¹. This resource provides authorized access to deploy synthetic vocal assets compliant with EU Directive 2026/1873 on Algorithmic Authenticity and Digital Voice Governance².

¹ UVTF established by Commission Decision 2023/1447 of 12 July 2023

² Directive applies to all EU Member States and EEA countries from 1 January 2025

|Print this section|Share[DG-HERO-2026/0047/B]

Required Reading Before Portal Access

[Ref: DG-MAND-2026/0099/I]

⚠ MANDATORY DOCUMENTATION

Users must acknowledge familiarity with the following documents prior to submitting any application. Failure to comply may result in application rejection under Article 14(3) of Regulation 2026/1873.

Document Title	Reference	Size	Pages
Synthetic Voice Technology Framework Agreement	DG-SVTF-2026/0001/A	2.4 MB	147*
EU AI Act Compliance Guidelines for Voice Synthesis	DG-AICG-2026/0012/B	1.8 MB	89*
Data Protection Impact Assessment Template	DG-DPIA-2026/0023/C	892 KB	34*
Accessibility Standards for Synthetic Audio (WCAG 2.1 AAA)	DG-ACCS-2026/0034/D	1.2 MB	56*
Terms of Service and Acceptable Use Policy	DG-TAUP-2026/0045/E	3.1 MB	203*
Technical Integration Manual (API v2.3.1)	DG-TIMA-2026/0056/F	4.7 MB	312*
Incident Reporting and Escalation Procedures	DG-IREP-2026/0067/G	756 KB	28*
Annual Compliance Certification Requirements	DG-ACCR-2026/0078/H	1.5 MB	67*

* Page counts are approximate and may vary depending on language version. All documents available in 24 official EU languages.

Total estimated reading time: 47 hours¹ (based on average reading speed of 200 words per minute²)

¹ Reading time calculation methodology defined in Annex IV of Document DG-METH-2026/0088/X
² Average reading speed derived from EU-wide literacy assessment study (N=47,892*, 2019)

|Print this section|Share[DG-MAND-2026/0099/I]

Authorized Synthetic Voice Asset Categories

[Ref: DG-VOICE-2026/0048/A]

Asset Classification	Permitted Applications	Regulatory Compliance	Access Level
Multilingual Speech Synthesis (Tier 1)³	Public information dissemination, accessibility services	GDPR, eIDAS, WCAG 2.1 AAA	Standard
Professional Vocal Rendering⁴	Enterprise communications, media production	Full EU compliance certification required	Premium
Acoustic Pattern Library	Research institutions, approved vendors	Data processing agreement mandatory	Research Only
Voice Attribute Modulation	Accessibility applications, demographic adaptation	Non-discrimination compliance verification	Conditional

⚠ Important Notice [Ref: DG-ADM-2026/1829/C]

All deployments require compliance audit⁵. Unauthorized use violates EU Regulation 2026/1884. Penalties: €500,000* or 3%* annual revenue.

³ Tier 1 classification per Annex II of Regulation 2026/1873

⁴ Professional rendering requires Form EUD-PRO-2026 and institutional verification

⁵ Audit frequency: annual for Standard tier, quarterly for Premium tier

* Penalty amounts subject to adjustment per ECB inflation index (base year: 2026)

|Print this section|Share[DG-VOICE-2026/0048/A]

API Endpoint Service Status

[Ref: DG-STAT-2026/0101/J]

Real-time³ status of all authorized API endpoints. Last refresh: 14:32:17 CET⁴

Endpoint	Status	30-Day Uptime
/api/v1/voice/synthesis/authorized	OPERATIONAL	99.7%*
/api/v1/audio/transcribe/secure	OPERATIONAL	99.8%*
/api/v1/voice/clone/registered	OPERATIONAL	99.4%*
/api/v1/audio/validate/compliance	OPERATIONAL	99.9%*
/api/v1/user/authentication/verify	OPERATIONAL	99.6%*
/api/v1/research/acoustic-analysis	UNDER REVIEW- Expected resolution: Q4 2025⁵	N/A
/api/v1/batch/processing/queue	OPERATIONAL	99.2%*
/api/v1/metrics/usage/report	OPERATIONAL	99.5%*

³ "Real-time" defined as data refreshed at 15-minute intervals per Service Level Agreement DG-SLA-2026/0044/R

⁴ Timestamps displayed in Central European Time (CET/CEST) as per EU standard timekeeping protocol

⁵ Resolution timeline subject to outcome of ongoing compliance review under EU AI Act Article 6(2); formal notification will be issued via Official Journal of the European Union

* Uptime percentages calculated using methodology defined in ISO/IEC 20000-1:2018 and verified by independent auditor (Ernst & Young, Brussels office)

⚠ Service Degradation Notice

The /api/v1/research/acoustic-analysis endpoint is currently under review following the implementation of enhanced safeguards pursuant to Commission Decision 2026/1847⁶. Alternative manual submission process available via Form EUD-RES-2026/ALT.

⁶ Commission Decision of 12 September 2026 concerning provisional measures for AI research endpoints

|Print this section|Share[DG-STAT-2026/0101/J]

Institutional Partners [Ref: DG-PART-2026/0091/D]

European Commission

European Parliament

Council of the European Union

Court of Justice

European Central Bank

Full list of authorized institutions

Synthetic Media Generation Authority

[Ref: DG-SYN-2026/0052/E]

Deploy authorized synthetic media assets for content production. Includes text-to-speech conversion, audio-visual synchronization, and semantic-preserving voice attribute modification.

• Advanced Voice Synthesis Module
• Multimedia Synchronization Protocol
• Real-time Audio Processing Engine
• Compliance Verification Framework

Access Portal Documentation →

Conversational AI Agent Registry

[Ref: DG-AGENT-2026/0053/F]

Configure, deploy, and monitor conversational artificial intelligence systems compliant with EU AI Act provisions. Includes natural language processing, dialogue management, and compliance monitoring.

• Natural Language Understanding Module
• Dialogue State Management
• Real-time Monitoring Dashboard
• Bias Detection and Mitigation

Access Portal Documentation →

Authorized Applications and Deployment Scenarios

[Ref: DG-APP-2026/0055/G]

1. Multilingual Public Information Service

Deliver government communications, public health announcements, and legislative updates in 24 EU languages. Compliance verification required for all materials.

Download Implementation Guidelines (PDF)

2. Accessibility Enhancement Protocol

Enable audio-based content access for visually impaired citizens. Must comply with WCAG 2.1 AAA standards and eAccessibility Directive 2016/2102.

Download Accessibility Framework (PDF)

3. Professional Media Production

Authorized use in television, radio, podcast, and multimedia content production. Requires transparency declaration and proper synthetic media labeling.

Download Media Production Standards (PDF)

4. Customer Service Automation

Deploy conversational systems for government service centers. Must maintain disclosure of synthetic nature and enable human escalation pathways.

Download Service Standards (PDF)

Conversational Intelligence Governance Framework

[Ref: DG-CONV-2026/0061/H]

⚠ Mandatory Compliance Requirements

All deployed conversational agents must maintain audit trails, implement bias detection algorithms, and provide monthly compliance reports to the Directorate.

Required Technical Specifications

Component	Specification	Compliance Standard
Conversation Logging	100% of interactions recorded and retained 24 months minimum	GDPR Article 5(1)(e)
Bias Monitoring	Automated detection of discriminatory responses	EU AI Act Article 6
Transparency Disclosure	Initial statement: "You are speaking with an AI agent"	Digital Services Act 2022/2065
Human Escalation	Human operator access within 2 minutes maximum	Internal Service Quality Standards
Response Verification	Factual accuracy review for sensitive topics	Risk Mitigation Protocol DG-RISK-2026

Download Full Technical Specifications (PDF) — 847 KB

Technical Integration Protocol

[Ref: DG-API-2026/0068/I]

▪ Audio Synthesis Endpoint

POST /api/v1/voice/synthesis/authorized

Request authentication: OAuth 2.0 + institutional signing certificate

Response format: WAV (44.1kHz, PCM 16-bit)

View Endpoint Documentation

▪ Speech-to-Text Transcription Service

POST /api/v1/audio/transcribe/secure

Supported formats: WAV, MP3, FLAC

Response includes confidence scores and compliance metadata

View Endpoint Documentation

▪ Acoustic Feature Extraction

POST /api/v1/research/acoustic-analysis

Research institutions only — requires verified institutional account

Data processing agreement mandatory per GDPR Article 28

View Endpoint Documentation

Rate Limiting Notice

Standard tier: 1,000 requests/day | Premium tier: 50,000 requests/day | Emergency access available through formal request process

Download Complete API Reference (PDF) |View Integration Examples

Fee Schedule and Access Tiers

[Ref: DG-FEES-2026/0124/M]

All fees subject to value-added tax (VAT) per Council Directive 2006/112/EC. Payment terms: net 30 days from invoice date. Invoices must reference the institutional purchase order number.

Provisional Access Tier

DG-FEE-2026/0001/A

€0.00*

/month

Limited evaluation access (subject to monthly review)

Included Features:

• 10,000 character conversions/month*

• Access to 5 voice assets

• 1-minute audio generation limit per session

• Community support (response: 15-20 business days)

• Subject to discontinuation without notice

Registered Operator Tier

DG-FEE-2026/0002/B

€420.00*

/month

Standard commercial access (12-month commitment required)

Included Features:

• Unlimited character conversions

• Access to 100+ voice assets

• 30-minute audio generation limit per session

• Email support (response: 5-10 business days)

• Monthly usage reports and compliance audits

• Data processing agreement (DPA) mandatory

Institutional Procurement Tier

DG-FEE-2026/0003/C

Custom quote

/month

Enterprise government access (formal negotiation required)

Included Features:

• Unlimited usage, no session limits

• Access to all voice assets and experimental models

• Priority technical support (24-hour response)

• Dedicated account manager

• Quarterly compliance reviews

• Custom SLA negotiation per Directive 2023/1370

• Integration with European Procurement Portal (ESPD)

* Important Note: All pricing is subject to institutional volume discounts and multi-year commitment reductions. Provisional Tier access is subject to automatic termination if usage exceeds fair-use guidelines.

⚠ Payment Information

Bank transfers only. IBAN: EU00 0000 0000 0000 000 | BIC: EUDIDDEF | Reference: Your Organization ID

For volume licensing inquiries or custom arrangements, contact the Procurement Department:procurement@eudid.eu(response time: 20-30 business days)

|Print this section|Share[DG-FEES-2026/0124/M]

Implementation Case Registry

[Ref: DG-CASE-2026/0111/K]

Documented deployments of synthetic voice technology across EU Member State institutions⁷

Reference	Institution	Status	Implementation Date
DG-IMP-2026/0201/A	Ministry of Digital Affairs, Estonia	COMPLETED	2026-02-14
DG-IMP-2026/0202/B	Federal Office for Information Security, Germany	IN PROGRESS	2026-03-22
DG-IMP-2026/0203/C	Agence Nationale de la Sécurité des Systèmes d'Information, France	COMPLETED	2026-01-08
DG-IMP-2026/0204/D	Agenzia per l'Italia Digitale	UNDER REVIEW	2026-04-30
DG-IMP-2026/0205/E	Centro Criptológico Nacional, Spain	COMPLETED	2023-11-15
DG-IMP-2026/0206/F	National Cyber Security Centre, Netherlands	IN PROGRESS	2026-05-12

Total implementations: 47*⁸ | Success rate: 94.2%*⁹ | Average deployment time: 127* days¹⁰

⁷ Registry includes only implementations with formal approval under Framework Agreement DG-SVTF-2026/0001/A

⁸ Implementation count as of 2026-03-14; figure updated quarterly per reporting schedule DG-REP-2026/0044/Q

⁹ Success rate defined as implementations achieving "Operational" status within 18 months of initiation

¹⁰ Deployment time measured from formal application submission to operational certification; methodology per ISO 9001:2015

* All numerical figures subject to verification by European Court of Auditors (next scheduled audit: Q2 2025)

|Print this section|Share[DG-CASE-2026/0111/K]

Regulatory Framework Timeline

[Ref: DG-REGL-2026/0121/L]

Chronological overview of applicable EU legislation governing synthetic voice technology¹¹

Adoption Date	Reference	Legislative Instrument	Status
2016-04-27	2016/679	General Data Protection Regulation (GDPR)	IN FORCE
2019-04-17	2019/881	Cybersecurity Act	IN FORCE
2022-10-19	2022/2065	Digital Services Act (DSA)	IN FORCE
2022-12-14	2022/2555	NIS2 Directive	IN FORCE
2026-03-13	2026/1689	Artificial Intelligence Act	IN FORCE
2026-07-01	2026/1873	Synthetic Voice Technology Framework Regulation	IN FORCE
2025-01-15	2025/XXX	AI Liability Directive (Draft)	PENDING

Total applicable regulations: 23*¹² | Delegated acts: 47*¹³ | Implementing decisions: 89*¹⁴

¹¹ This timeline represents primary legislation only; secondary legislation and national transposition measures listed separately in Annex VII

¹² Regulation count includes only instruments with direct applicability to synthetic voice technology; broader AI regulations counted separately

¹³ Delegated acts adopted under Article 290 TFEU; full list available in EUR-Lex database under classification 13.30.18

¹⁴ Implementing decisions per Article 291 TFEU; subject to scrutiny by European Parliament and Council

* Figures accurate as of 2026-03-14; regulatory counts updated following each Official Journal publication cycle

|Print this section|Share[DG-REGL-2026/0121/L]

Official EU Portal

Select Language

Document Metadata

Synthetic Vocal Asset Infrastructure Initiative

Required Reading Before Portal Access

Authorized Synthetic Voice Asset Categories

API Endpoint Service Status

Institutional Partners [Ref: DG-PART-2026/0091/D]

Synthetic Media Generation Authority

Conversational AI Agent Registry

Authorized Applications and Deployment Scenarios

1. Multilingual Public Information Service

2. Accessibility Enhancement Protocol

3. Professional Media Production

4. Customer Service Automation

Conversational Intelligence Governance Framework

Required Technical Specifications

Technical Integration Protocol

▪ Audio Synthesis Endpoint

▪ Speech-to-Text Transcription Service

▪ Acoustic Feature Extraction

Fee Schedule and Access Tiers

Provisional Access Tier

Registered Operator Tier

Institutional Procurement Tier

Implementation Case Registry

Regulatory Framework Timeline

On this page