ZTABS builds AI voice agents for Doha enterprises in telecommunications, banking, government services, healthcare and hospitality. Qatar's service culture sets high customer service expectations, and voice remains the preferred channel for complex enquiries. Gulf Arabic dialect understanding with Arabic-English code-switching is the critical technical requirement for Doha voice agents. ZTABS is remote-first by design from our Houston, Texas office.
Doha voice agent demand spans telecommunications customer service (Ooredoo, Vodafone Qatar), banking account management and card services (QNB, QIB, Commercial Bank), government citizen services (Hukoomi portal and ministry lines), healthcare appointment booking (Hamad Medical Corporation and private hospitals), and hospitality concierge and reservation services. Our AI voice agent development services cover enterprise voice automation across markets.
AI voice agent development in Doha at a glance
| Need | What we build | Key requirement |
|---|
| Telecom | Service enquiries, plan changes, recharge | Gulf Arabic, bilingual |
| Banking | Account services, card blocking, payments | QCB, identity verification |
| Government | Citizen enquiries, appointment booking | Arabic, Hukoomi integration |
| Healthcare | Appointment booking, prescription enquiries | Arabic medical terminology |
| Hospitality | Concierge, reservations | Multilingual guests |
How voice agents work
A voice agent receives or places calls through telephony infrastructure. Audio converted to text by STT. Text processed by language model. Response converted to speech by TTS.
STT selection. Provider and model evaluated for Arabic including Gulf Arabic dialect. Gulf Arabic has unique expressions and pronunciation patterns distinct from MSA. Arabic-English code-switching is common in Doha ("I want to change my plan" mixed with Arabic phrases within the same utterance). STT must handle this bilingual input. Evaluation uses held-out test sets with Gulf Arabic speakers, including code-switching samples, with baseline, error categories and median, P95 and P99 metrics.
TTS selection. Arabic neural TTS evaluated for natural pronunciation, intonation and speaking rhythm appropriate for Qatari service context. Voice selection matches brand personality. Arabic-Indic numeral handling for amounts and dates.
Model selection. Language model for the specific task. Telecom service enquiries need different model behaviour than banking. Gulf Arabic expressions and conversational patterns evaluated. Model size balanced against latency.
Building a voice agent for Doha telecom or banking?
Tell us about your call volume, use case and compliance requirements. We start with assessment covering telephony, Arabic language evaluation and regulatory mapping.
Inbound versus outbound
Inbound voice agents answer calls. Risk-approved identity verification before revealing protected information. Handle within the owner-approved action envelope. Transfer to human when exceeding scope.
Outbound voice agents place calls for reminders, notifications and follow-ups. Qatar marketing call regulations assessed per actual outbound use case and entity scope. Service notifications distinguished from marketing. Outbound communication uses respectful, culturally appropriate tone.
Conversational design
Identity and verification. Before revealing account details, banking information or service status, verify caller identity through risk-approved methods. CLI, phone number, date-of-birth or voiceprint alone are not adequate authentication for protected banking actions. No secret PIN, password or OTP captured into audio, transcript, model or logs. Match verification level to action risk.
Gulf Arabic conversation. Arabic conversation conventions include appropriate greetings and culturally expected conversational flow. The voice agent follows Arabic conversation norms without jumping directly to business. Gulf Arabic expressions handled where the STT model supports them. Code-switching between Arabic and English handled where the agreed language scope supports it.
Arabic-English bilingual. Evaluated for Gulf Arabic and English with held-out test sets per agreed language scope. Language detected within initial speech and routed to the appropriate conversational path. Code-switching handling assessed per available STT provider coverage and actual caller demographics.
Repair and confirmation. Low-confidence triggers confirmation. Read-back of amounts (in QAR), dates, reference codes. Arabic-Indic numerals handled naturally. Explicit confirmation before consequential actions.
AI identity and recording disclosure. The voice agent identifies itself as an automated system at call start and offers a human alternative. Recording disclosure and consent follow immediately. Callers who decline stored recording cannot have audio capture or live STT silently continue; architecture provides a lawful alternate path.
Latency. Response time varies by the STT, model and TTS provider chain. Measured median, P95 and P99 end-to-end latency on agreed scenarios.
Qatar data protection and recording consent
Qatar's Personal Data Privacy Protection Law (PDPPL, Law No. 13 of 2016) governs personal data processing. NCSA (National Cyber Security Agency) requires prior permission for special nature data processing (Article 16). Recording consent and marketing call rules assessed per actual engagement scope, entity type and applicable regulator. CRA (Communications Regulatory Authority) has jurisdiction over telecommunications. English translations of Qatar law are expressly not the official legal version; the Arabic text governs. The agreed consent-first architecture discloses recording at call start. Callers who decline cannot have audio capture or live STT silently continue; architecture provides a lawful alternate path. No automatic reuse of customer recordings for after-session storage or continuous learning without agreed lawful purpose and rights.
Safe bounded workflow
Low confidence. Transfer to human when intent unclear.
Unsupported intents. Acknowledge, explain limitation, offer transfer.
Refusal and opt-out. Human transfer through staffed queue during agreed service hours. No forced bot interaction.
Caller distress. Emergency callers directed to Qatar emergency services (999). Voice agent is not emergency service.
Unknown outcome. Reconcile current remote object state and authoritative local ledger before retry. Bounded retries with limits and manual exception path. Receiver-side deduplication and atomic local updates.
Telecommunications voice agents
Qatar's telecom operators handle significant customer service volumes.
Service enquiries and plan management. Balance checking, plan changes, data bundle purchases and recharge. The voice agent understands Gulf Arabic expressions for these requests, identifies the current plan and presents options. Processing changes through operator-approved API integration with tenant and object-level authorisation.
Technical support triage. Basic troubleshooting for internet, connectivity and device issues. Guiding customers through diagnostic steps in Arabic and English. Escalation to specialised technicians for unresolved issues.
SIM management. SIM blocking for lost or stolen phones as a high-urgency operation. Identity verification, immediate SIM blocking through operator-approved integration and replacement scheduling. Available during agreed service hours including after-hours for emergency operations where the operator's programme supports it.
Banking voice agents
Qatar's financial institutions handle significant customer service call volumes.
Account services. Balance enquiries, recent transactions, account information. Identity verification through the bank's approved multi-factor methods per QCB requirements. Integration with core banking through bank-approved APIs.
Card services. Card blocking, activation, PIN reset, spending limit changes. Card blocking as immediate resolution. Multi-factor verification for consequential actions.
Payment reminders. Outbound voice agents for upcoming or overdue payment notifications. Respectful, culturally appropriate Arabic tone. No mandatory credential capture during outbound calls.
Government services voice agents
Qatari government services through Hukoomi and ministry lines.
Citizen enquiries. Information about government services, required documents and procedures. Arabic response in clear MSA or Gulf Arabic depending on caller preference. Routing to appropriate ministry or department based on enquiry content.
Appointment booking. Government service appointments for visa services, civil ID and municipality services. Write permissions require the government entity's approved programme.
Healthcare voice agents
Hamad Medical Corporation and private hospitals.
Appointment booking. Understanding patient needs in Arabic, checking availability, booking appointments through hospital-approved APIs. Arabic medical terminology for department routing. Booking confirmed after authoritative system acknowledgment.
Prescription enquiries. Prescription refill requests and medication enquiries through hospital-approved systems with tenant and object-level authorisation.
Hospitality voice agents
Qatar's hotel and service industry.
Concierge and reservations. Multilingual (Arabic, English and additional guest languages). Restaurant reservations, transport arrangements, activity booking. Integration with hotel systems through operator-approved APIs.
Integration requirements
Telecom systems. CRM, billing, provisioning, network diagnostics through operator-approved APIs.
Banking systems. Core banking, CRM through bank-approved APIs per QCB requirements.
Government systems. Hukoomi and ministry systems through government-approved APIs.
Telephony. SIP trunking, Qatari numbers (+974), concurrent capacity. Qatar telco provisioning programme and write permissions per operator's terms.
Voice-specific cost drivers
Voice agent development follows the custom software cost guide as general platform engineering context.
| Cost driver | Description |
|---|
| Telephony infrastructure | SIP trunking, Qatar numbers, concurrent capacity |
| STT provider usage | Per-minute by provider, model and language (Arabic) |
| TTS provider usage | Per-character or per-minute by voice and language |
| Model inference | Per-token pricing, scales with volume |
| Arabic language engineering | Gulf Arabic dialect handling, code-switching |
| Recording and storage | Consent-gated configured coverage, transcript storage |
| Custom voice | Branded AI voice per organisation (optional) |
Telephony, STT, TTS, LLM inference, minutes, concurrency and retention are separately scoped per engagement without invented operating quotes. Custom voice requires actual speaker consent and usage licence.
Illustrative Example
Hypothetical: Doha telecom customer service voice agent
A Qatari telecommunications provider handling hundreds of thousands of monthly calls across billing, plan management, technical support and SIM services in Arabic and English.
A voice agent would answer inbound calls, detect language (Arabic or English), verify customer identity through the operator's approved methods, handle routine enquiries (balance, plan information, data bundles, recharge), process plan changes through operator-approved APIs and route complex issues to specialised agents with full context. SIM blocking available as high-urgency operation. Gulf Arabic conversation conventions maintained. Recording consent disclosed at call start.
Assumptions: operator API access confirmed with CRM, billing and provisioning integration scoped, service scope documented with PDPPL compliance assessed, agreed consent-first recording architecture, telephony supports SIP with Qatari numbers, and Gulf Arabic STT and TTS evaluated with held-out test sets. Residual risks include Gulf Arabic ASR accuracy variation across speakers and noise conditions, code-switching recognition edge cases, system integration latency during peak periods and caller preference for human agents. No guarantee of specific automation rate, wait time elimination or satisfaction outcome.
This is a hypothetical scenario for illustration. Actual scope, timeline and cost depend on your requirements.
Cost and timeline
Voice agent development as general custom platform engineering uses four guide tiers.
| Scope | Cost range | Timeline |
|---|
| Internal Tool | $25K to $60K | 6 to 14 weeks |
| Business Application | $60K to $150K | 14 to 28 weeks |
| Enterprise Platform | $150K to $300K | 28 to 40 weeks |
| Large-Scale System | $300K to $500K+ | 40 to 60+ weeks |
2026 US vendor benchmark
Ranges are 2026 US-buyer guide-based planning benchmarks for a mid-complexity business application: multi-role access, 3 to 5 integrations, custom workflows and a modern web frontend.
| Vendor type | Typical cost | Timeline |
|---|
| Freelancer or dev collective | $20K to $75K | 4 to 10 months |
| Offshore agency | $40K to $120K | 5 to 11 months |
| Nearshore agency | $70K to $200K | 4 to 8 months |
| US agency (ZTABS tier) | $90K to $280K | 4 to 8 months |
| Big 4 or systems integrator | $250K to $800K+ | 8 to 16 months |
Ranges are 2026 US-buyer guide-based planning benchmarks for the scoped business-application comparison. Where applicable to the agreed engagement, legacy-system integration (SAP, Oracle or mainframe interfaces) and compliance engineering or readiness preparation (SOC 2, HIPAA safeguards or PCI DSS controls) carry an additional guide allowance of $20K to $80K across vendor paths. Independent CPA SOC 2 examination and opinion are scoped and priced separately from internal readiness engineering. The guide uses 15 to 25 percent of build cost annually as an ongoing-maintenance planning estimate; actual support scope and fees are agreed separately. These are scoped planning allowances, not an empirical market survey or voice-package rate card.
Need a scoped estimate for Doha voice agent development?
Share your use case, call volume and compliance requirements. We provide a scoped estimate after reviewing both four-tier scope and vendor benchmark context.
Proposal evaluation criteria
- Relevant voice workflow experience and customer-approved action authority with supporting evidence
- Engineering and integration reliability including unknown-outcome recovery, bounded retries and manual exception paths
- Assessed privacy, security and regulatory boundaries with clear independent assurance responsibilities
- Speech, language and accessibility evaluation with held-out test sets and agreed acceptance criteria
- Transparent scope, milestones, change control, operating fees and ongoing support terms
- Human handover, training, validation and support with actual customer and provider responsibility allocation, agreed staff and user languages, and documentation and translation responsibilities
Qatar regulation
PDPPL. Personal Data Privacy Protection Law (Law No. 13 of 2016) governs personal data processing in Qatar. Special nature data (Article 16) requires NCSA prior permission. English translations are expressly not the official legal version.
QCB. Qatar Central Bank regulates financial services. Banking voice agents comply with QCB requirements per institution type and service scope.
CRA. Communications Regulatory Authority has jurisdiction over telecommunications. Telecom voice agents comply per applicable CRA requirements.
Recording consent. Assessed per actual legal framework, entity type and engagement scope. Consent-first design at call start. Lawful alternate path for refusal.
How delivery works
ZTABS is remote-first by design from our Houston, Texas office. Houston is Central time; Doha is Arabia Standard Time (AST, UTC+3), usually 9 hours ahead and 8 hours during US daylight saving time. Houston morning overlaps with Doha evening. Asynchronous coordination for non-overlapping hours.
When ZTABS is not the right fit
- Simple IVR tree with no AI for this engagement. Traditional IVR vendors handle press-1 menus.
- On-site embedded team for this engagement. ZTABS is remote-first by design.
- Arabic speech model research for this engagement. Voice agent delivery uses evaluated STT providers; fundamental Gulf Arabic research is separately scoped.
- Clinical decision support for this engagement. Routing to clinicians, not autonomous decisions.
- QCB regulatory consulting for this engagement. Compliance engineering is part of delivery; independent regulatory advice is separately agreed.
White-label AI products
ZTABS builds custom white-label AI products for clients to brand and resell. Doha enterprises can commission white-label voice AI platforms for distribution.
Evaluation methodology
Representative test calls. Expected uses, edge cases, accents, noise, channel quality.
Gulf Arabic test sets. Gulf Arabic speakers, code-switching samples, MSA comparison. STT accuracy on agreed test set with baseline, error categories and median, P95 and P99 metrics.
English evaluation. English speech recognition evaluated separately per agreed scope.
Cultural review. Greeting conventions, conversation flow and response tone reviewed for Qatari cultural appropriateness.
Task completion. Intent identified and action completed. Human baseline comparison.
Data governance
Scope authorised tenant and object data access and actual provider product, API, programme, version and licence. Retrieval and writes are separate. Permission changes propagate across indexes, embeddings, caches, logs and fine-tunes as applicable with disclosed residual lag and mitigations. No automatic reuse of recordings, customer research, employee chat, proprietary code or licensed publications for training. No blanket after-session storage or continuous learning reuse of customer recordings without agreed lawful purpose and rights. Agreed data-processing terms, model configuration, retention, deletion, legal holds and approved test data per engagement.
Monitoring and operations
Real-time dashboards. Call volume, completion rates, transfer rates, STT accuracy, latency. Per-language metrics.
Model updates. Tested in staging. A/B testing. Rollback. Financial and government model updates reviewed by designated authority.
Recording and transcription. Permitted coverage per consent architecture. Minimisation, access controls, retention, deletion and legal hold per engagement.
Handover
Voice agent handover includes runbooks, telephony configuration, model versions, provider accounts, monitoring dashboards, evaluation methodology, Arabic consent architecture and escalation procedures.
Ready to build a Doha voice agent?
Share your call volume, use case and compliance requirements. We start with assessment covering telephony, Arabic language evaluation and Qatar regulatory mapping.