AI Launches: Conversational AI, Cybersecurity, Wellness & Communication

The Tech Panda takes a look at recent launches in the superfast field of Artificial Intelligence (AI).

Conversational AI: Industry-First Customer Accent Conversion to Call Centers to Reduce Agent Cognitive Load

Krisp announced customer-side Accent Conversion, a real-time, inbound Accent AI capability designed to improve agents’ comprehension during live customer calls. The launch addresses a persistent challenge in customer experience: even when audio is clear, comprehension can break down across customers’ accents and dialects, increasing effort for agents and friction for customers.

“In contact centers, the strain isn’t just on customers; it’s on agents who are often working in a second language and processing multiple accents all day,” said Davit Baghdasaryan, CEO and Co-Founder of Krisp. “When agents have to work harder just to understand what’s being said, cognitive load increases and performance suffers. Customer Accent Conversion reduces that effort in real time, helping agents stay focused, resolve issues faster, and deliver better customer experiences without asking customers to repeat themselves or change how they speak. The downstream impact is tangible — reduced handle time, faster resolutions, and improved satisfaction on both sides of the call.”

FEATURES

• With agent-side accent conversion already in market, Customer Accent Conversion extends real-time clarity to both sides of the call.
• Processing happens locally on the agent’s device, and raw audio is never stored or routed through external servers.
• Operates as a virtual microphone and speaker, enabling deployment across major CCaaS platforms and softphone applications without workflow changes.
• For contact centers, the impact is operational: reductions in repetition and handle time, alongside double-digit improvements in agent and customer satisfaction.

Cybersecurity: Enterprise Application Security for the AI & Post-Quantum Era

At AppWorld, F5 (NASDAQ: FFIV), the cybersecurity company, unveiled new security capabilities that strengthen protection for AI-driven and modern applications. Integrated within the F5 Application Delivery and Security Platform (ADSP), the innovations unify intelligent threat protection, zero trust access controls, and crypto-agile architecture to help enterprises secure distributed applications and prepare for emerging post-quantum risks.

“Security teams do not need more alarms. They need fewer gaps,” said Kunal Anand, Chief Product Officer at F5. “ADSP closes the loop from finding risk to enforcing protection. That includes moving from identified AI model vulnerabilities to validated runtime guardrails, AI-powered risk scoring, and a practical path to zero trust and post-quantum readiness. The point is simple: move faster while reducing your threat landscape.”

FEATURES

• New to the F5 ADSP, F5 AI Remediate closes a gap between identifying AI model vulnerabilities with F5 AI Red Team and enforcing validated runtime protections with F5 AI Guardrails. It accelerates time to response by automating the creation, optimization, and validation of targeted guardrail packages, enabling security teams to put evidence-backed protections into production with human approval. As part of the ADSP, it helps organizations reduce exposure quickly without disrupting live AI applications, eliminating repetitive tasks while supporting accountable AI adoption.
• F5 Distributed Cloud WAF release introduces AI-powered risk scoring, transforming manual processes into automated protections. With outcome-based blocking policies, CISOs, SecOps, and NetOps teams can eliminate threats, reduce operational burden, and achieve rapid time to protection, all while maintaining low false positive rates through layered analysis.
• F5 Web Application and API Protection (WAAP) solutions such as this minimize reliance on signature-level exceptions and tedious tuning for enterprises with significant application portfolios. Whether safeguarding on-premises hardware or virtual systems, cloud-based SaaS applications, or containerized deployments, F5 enables consistent, enterprise-grade security across ecosystems.

Wellness: India’s First AI-Powered Beauty Intelligence Platform to Eliminate Skincare Guesswork

Style Lounge, a beauty technology platform, launched its “Beauty Intelligence Platform,” an innovative ecosystem that bridges the gap between artificial intelligence and dermatological science. By moving beyond simple salon bookings, Style Lounge is introducing a data-driven approach to beauty that replaces influencer-led guesswork with measurable skin diagnostics.

“At Style Lounge, we believe that the future of beauty must be diagnostic,” said Deepak Gupta, Co-Founder. “Most consumers today treat skin symptoms rather than root causes, often worsening their condition with mismatched products. Our AI platform transforms the beauty journey from subjective opinion into a precise, data-driven science. By diagnosing first, we ensure every treatment and product recommendation is as unique as the user’s skin.”

FEATURES

• Its AI-Powered Skin Analysis tool analyzes a simple smartphone selfie, and the system utilizes advanced computer vision and deep-learning models trained on different datasets to evaluate ten distinct skin parameters—including acne, hydration, pigmentation, and structural laxity. This technology detects micro-contrast variations and texture depth patterns invisible to the naked eye, providing users with objective skin scores and early warning indicators within seconds.
• The platform’s Intelligent Recommendation Engine then maps these diagnostic results directly to curated salon treatments and routines. Whether a user requires hyaluronic acid-based therapies for low hydration or targeted collagen-boosting treatments for enlarged pores, Style Lounge ensures a “Data ? Diagnosis ? Decision” workflow that reduces product misuse and promotes long-term skin health.
• Integrated with an AI Beauty Assistant and a seamless booking system, Style Lounge provides a comprehensive ecosystem where users can discover verified salons, compare treatments, and receive guaranteed service fulfillment based on their unique biological needs.

Communication: Enterprise Agentic AI Platform to Orchestrate Workflows Across Collaboration & Customer Experience

Zoom Communications, Inc. (NASDAQ: ZM) announced the expansion of its enterprise agentic AI platform, introducing new workflow orchestration capabilities across Zoom Workplace, Zoom Phone, and Zoom CX. The updates extend Zoom’s system of action for modern work, enabling organizations to automate tasks, trigger cross-system workflows, and turn meetings, calls, and customer interactions into completed business outcomes.

“The next phase of enterprise AI will be defined by the ability to move from conversation to action,” said Velchamy Sankarlingam, president of Product & Engineering at Zoom. “Zoom’s agentic AI platform is designed to orchestrate action across systems, turning every meeting, call, and customer interaction into a trigger for workflow automation.”

FEATURES

• Expanding AI Companion 3.0 more broadly across the Zoom platform and introducing custom AI agents to orchestrate workflows in Zoom and third-party systems
• Custom and prebuilt AI agents with no-code orchestration
• New third-party integrations for AI Companion, including 10 secure enterprise search connectors
• New AI-first canvases (Zoom AI Docs, AI Sheets, and AI Slides)
• Zoom Phone Mobile
• SMS for Zoom Virtual Agent AI Receptionist
• AI Expert Assist 3.0 for Zoom Contact Center
• Natural-language customer workflow orchestration
• Meeting security enhancements

Conversational AI: The World’s First Open Full-Duplex Conversational AI Model in Hindi

Josh Talks, a company building large-scale speech datasets and evaluation infrastructure for AI, announced the development of the world’s first full-duplex conversational AI model in Hindi, an AI system capable of listening and speaking at the same time, enabling voice interactions that more closely resemble natural human conversation.

Shobhit Banga, Co-Founder, Josh Talks, said, “Voice AI has learned to recognise speech and generate speech. What it has not yet learned well is how to participate in conversation. This research shows that when models are trained on large-scale natural dialogue, they can begin to learn the rhythm of how people actually speak to each other. By building this infrastructure using Hindi conversations, we have demonstrated that full-duplex conversational systems can be developed beyond English. The mission of Josh Talks AI is to make machines talk like humans. The 26,000-hour conversational dataset used for this work represents the first milestone in a broader effort to build a one-million-hour corpus of natural conversations, one of the largest datasets ever created for conversational voice AI.”

FEATURES

• The model can listen and speak simultaneously, enabling AI conversations that mirror how people actually talk.
• Trained on 26,000 hours of real Hindi conversations across 14,695 speakers, capturing interruptions, pauses and natural dialogue flow.
• Demonstrates how large-scale conversational datasets can unlock next-generation voice AI for Indian languages.
• Unlike traditional voice assistants that operate in a turn-based format, where users must finish speaking before the system responds, it can listen and speak simultaneously. This allows it to acknowledge users mid-sentence, respond in real time, and handle interruptions without breaking the flow of conversation.