How Businesses Can Use Sonic-3.6 for Voice and Audio Applications
Sonic-3.6 raises the bar for AI-generated voice quality, but using it in business demands sound compliance on consent, disclosure, and brand representation.
On August 27, 2026, Cartesia introduced Sonic-3.6, their latest AI text-to-speech (TTS) model. Sonic-3.6 produces synthetic voice outputs and was developed as a direct improvement over the earlier Sonic-3.5 model, aiming for higher naturalness and audio quality.
Sonic-3.6 differs from previous text-to-speech models—like standard TTS engines from Google, Amazon, or prior Cartesia versions—by being rated as more natural by listeners in up to 93% of blind tests across fifteen locales. This leap in preference rate makes its output more suitable for workflows that require humanlike audio, such as realistic phone agents or dynamic audio generation.
Firms using voice agents, phone-based support, or producing branded audio content can now consider workflows that previously required human voice talent. For regulated industries and customer-facing teams, this starts critical questions about consent, active disclosure, and managing synthetic voices in a way that aligns with both regulatory and reputational requirements.
What Sonic-3.6 Offers for Business Voice and Audio Workflows
Sonic-3.6 outputs highly natural AI-generated speech that listeners prefer significantly over previous versions, making it competitive for real-time phone agents, voiceovers, and audio content creation.
The model was benchmarked in blind head-to-head tests, where listeners chose Sonic-3.6 in up to 93% of trials across fifteen locales. This suggests strong generalization and quality across diverse linguistic and regional requirements.
Key business applications include:
• Phone-based customer service agents or IVR (interactive voice response) systems.
• Automated outbound calls for appointments, reminders, or notifications.
• Audio content for training, branded podcasts, and explainer videos.
• Voiceover for marketing, product tutorials, or internal communications.
- Natural, humanlike synthetic voices for client-facing calls.
- Fast, scalable voice generation for high-volume content.
- Support for multiple locales and accents for global teams.
First Month Free
Get one month of Starlink free when you sign up through this link. Fast, reliable internet at home and on the go.
Brand Voice and Identity: Maintaining Trust in Audio
Using Sonic-3.6 for business audio raises important brand-voice and identity considerations, as synthetic speech can sound nearly indistinguishable from a human operator.
Teams need clear protocols to define which types of interactions can use synthetic voices and which require authentic human contact. For instance, onboarding calls or sensitive support requests may merit direct human handling, while routine information or status updates might be appropriate for AI voices.
A consistent, transparent brand voice also means training the model (or selecting a preset) that matches your organization's tone and values. Companies should regularly review and update their synthetic voices to ensure alignment with evolving brand standards.
In our operational work helping firms implement voice agents, we observe that early brand-voice missteps—such as overly robotic output or inconsistent persona—can erode trust with both external customers and internal staff.
Consent, Disclosure, and Ethical Use of Synthetic Voices
Deploying Sonic-3.6 for voice calls or content requires careful handling of consent and disclosure, especially in regulated or consumer-facing settings.
Most jurisdictions require businesses to inform individuals when they are interacting with an AI or recorded voice rather than a live human. This may include up-front verbal disclosures at the start of calls, visible indicators in messaging workflows, or written notices on company websites.
Consent is also necessary when customer conversations are recorded or synthesized, which can intersect with state and federal call recording laws, data protection regulations, and industry standards (like HIPAA or GDPR for health-related communications).
Failing to disclose or obtain consent for synthetic voice use can result in customer complaints, regulatory penalties, or reputational risk.
- Announce synthetic or AI voice on calls before collecting personal data.
- Train staff and audit agent scripts for compliance with consent laws.
- Consult legal and compliance teams on required disclosures.
Example Business Workflows Using Sonic-3.6
Organizations are using Sonic-3.6 to streamline operations, improve reach, and reduce manual voice work in several practical ways.
Common workflows include:
1. Automated appointment confirmations and reminders for healthcare providers, with consent disclosures embedded in call scripts.
2. Financial-services onboarding workflows, where routine compliance notices are delivered by synthetic voice, with a clear announcement of the AI agent.
3. E-commerce post-purchase phone follow-ups, using preset brand voices and adapting messaging for each locale.
4. Training and e-learning modules produced at scale with humanlike voiceovers to maintain consistent tone across languages.
In the SMB implementations we have supported, adoption friction tends to center on regulatory ambiguity and disclosure mechanics, rather than on the underlying technical quality.
Managing Compliance and Risk with Sonic-3.6 Deployments
Meeting compliance standards when deploying Sonic-3.6 means systematizing documentation, audit, and ongoing review of both consent and voice output.
Key steps include:
• Maintaining an up-to-date inventory of workflows where AI voice is used.
• Requiring explicit user consent at the start of calls (or visits), recorded in CRM or workflow logs.
• Implementing regular audits of agent behavior and disclosure practices.
• Training all staff interacting with voice or audio agents on correct procedures and escalation points.
In heavily regulated industries, such as healthcare, finance, and law, involving compliance teams from the outset reduces costly rework and speeds time-to-value.
Deployment Checklist: Putting Sonic-3.6 to Work in Your Team
A successful Sonic-3.6 rollout involves technical setup, training, legal review, and operational monitoring.
Recommended checklist for business teams:
1. Define all use cases for synthetic voice—including outbound, inbound, and recorded content.
2. Review disclosure and consent requirements in your jurisdictions before live deployment.
3. Select or train on-brand voices, testing with real sample calls.
4. Add disclosure lines to all scripts and documentation.
5. Audit initial deployments for compliance and user experience issues.
6. Schedule regular review points to re-assess compliance risks, especially if regulations or model capabilities evolve.
- Map use cases by risk and compliance level.
- Document all disclosures made to users.
- Continue monitoring regulations and update practices as needed.
Frequently Asked Questions
- Sonic-3.6 was rated as more natural than its predecessor Sonic-3.5 in up to 93% of blind tests across fifteen locales, offering improved humanlike quality suitable for business audio workflows.
- Yes, but deployments must ensure compliance with industry standards like HIPAA or GDPR, as well as local consent and disclosure regulations for synthetic voice use.
- Most US states and international jurisdictions require up-front disclosure when an interaction is handled by an AI or synthetic voice, especially if collecting personal or sensitive information.
- Consent should be explicit, recorded, and reviewed periodically, following local laws on call recording and synthetic voice use.
- Failing to ensure proper disclosure and consent can lead to regulatory penalties, reputational damage, and loss of customer trust.
- Yes, unit tests for Sonic-3.6 showed preference for its output in fifteen different locales, suggesting strong multi-locale capability. The full language list should be confirmed in Cartesia’s documentation.
- Business teams should schedule regular audits, especially when workflows change or new regulations are enacted. Ongoing monitoring helps prevent compliance drift.
Book a Free AI Compliance Review
Deploying Sonic-3.6 for business audio or phone agents introduces compliance and brand risks along with new capabilities. Our team will review your workflow plans, look for regulatory gaps, and suggest practical controls for disclosure, consent, and audit. Book a free 30-minute consultation to map your rollout.
Book a Consultation