Inworld AI

Consumer AI Accelerator Week 3 recap! Featuring speakers from our growth and VC fundraising sessions: Alex Mashrabov (Higgsfield AI), Fai Nur (Status), Jean Wang (Inworld AI), and Jasper Brand (BITKRAFT Ventures). We discussed how consumer AI products actually get built, distributed, and financed. Swipe the carousel for the takeaways. Reminder: Demo Day · March 3 · San Francisco Funds including BITKRAFT Ventures, Lightspeed, Menlo Ventures, Maven, and others will be in the room. Seats are limited. RSVP via the link in the first comment.

We just expanded our voice library to include 44 new voices. The full library now covers every major use case, with voices purpose-built for the context they operate in, across multiple languages: - Companions: general-purpose voice assistants, relationship and social companions, fitness coaching, and personal growth companions - Enterprise workforce: customer support and CX, sales automation, internal knowledge Q&A, recruiting, and product and user research - Learning and education: language learning and tutoring, professional training and onboarding, and skill-building and coaching - Developer assistants: AI coding assistants, development agents, and code collaboration tools - Health and wellness: fitness and lifestyle coaching, mental health and spiritual support, wellness guidance, patient education and support, clinical assistance tools, and healthcare workflows - Interactive media: game characters and NPCs, IP-based experiences, interactive content, ads and avatars, and news, sports, and entertainment Picking the right voice for the right context makes a measurable difference in your user experience. Each voice in our library is designed with a specific use case in mind, not as a general-purpose option. You can browse the full library and listen to samples at inworld.ai/tts

"The first wave of AI made existing things cheaper. The next wave makes things that couldn't exist before." We were featured in Business Insider alongside our partners at Particle, Luvu, and Status for our vision of what comes next in AI. We call it the Second Wave. Most AI investment so far has gone toward automation and efficiency. That matters, but it only reshuffles value within existing businesses. The real economic potential of AI is in creating entirely new products and experiences that consumers want to pay for. That shift demands a new kind of infrastructure: realtime responses, support for millions of concurrent users, and deeply personalized experiences. Read the full piece from Alistair Barr: https://lnkd.in/gFKEsVhf

Week 3 recap of Consumer AI Accelerator. Last week, Avisha Mehra Naganath from Khosla Ventures, Stacey S. from Oyster®, and Lidiya Terpel from Inworld AI broke down how fast-growing startups actually build teams: • The “power of six” rule • Why your first 10 hires define the next 100 • When hiring CXOs is too early • How to recruit candidates stronger than you Below are 10 key takeaways on strategic hiring in a fast-scaling company. Our Demo Day’s happening March 3 in San Francisco! VC seats are limited, RSVP through the link in the first comment.

ICYMI: To celebrate the release of Inworld TTS 1.5, we're giving away up to $100 in promotional credits for Ultravox customers who run production agents with Inworld AI voices. Read more: https://lnkd.in/gtjpWaHP More details on how to qualify: https://lnkd.in/g5Z96H-s

Week 2 recap of Consumer AI Accelerator. This week we ran two sessions around one core question: What really matters when you’re building a consumer AI product that can scale? Session 1: Metrics that actually drive growth With Brian Yun, MD(Endo Health (a16z SR)), Sara Beykpour (Particle), and Florin Radu (Inworld AI), we talked about metrics beyond vanity numbers. Your North Star shouldn’t be looked at in a vacuum. If you optimize the wrong signals (downloads, surface-level engagement), growth can look healthy while the product slowly drifts away from real user value. Strong teams rely on dashboards that connect user behavior, retention, and unit economics, not isolated numbers. Session 2: Building for hundreds of millions of users With Alexander Chekan (Flo Health Inc.), we explored what it takes to scale consumer products to 400M+ users. We discussed why long-term defensibility comes from systems, not features. By systems, we mean things like: ➤ Feedback loops between user behavior and product decisions ➤ Data + personalization infrastructure ➤ Community, moderation, and trust mechanisms ➤ Distribution and retention systems that compound over time Features can be copied. Systems and communities are much harder to replicate. Our cohort keeps shipping, testing, and iterating fast, applying these ideas in real products. Demo Day’s happening March 3 in San Francisco! VC seats are limited, RSVP through the link in the first comment.

Inworld TTS 1.5, the fastest and highest-quality realtime voice AI models on the market, are now available on WaveSpeedAI. TTS 1.5 is the top-ranked model globally on both the Artificial Analysis and Hugging Face quality leaderboards, with ~200ms latency, greater stability and expressiveness, and 25x cost savings versus comparable alternatives. The TTS 1.5 family of models also includes multilingual support across 15 languages, high-quality instant and professional voice cloning with API support, timestamps and websockets, audio markups, and flexible deployment options including on-premises and on-device. We are excited to partner with WaveSpeedAI, a low-latency speech-to-speech platform for conversational AI applications. WaveSpeedAI's integration of TTS 1.5 enables developers to build real-time voice experiences with the quality, latency, and economics required to scale to millions of concurrent users. Learn more about TTS 1.5: https://lnkd.in/er3kwvuv Learn more about WaveSpeedAI: https://wavespeed.ai/

We now offer zero-data retention + HIPAA, SOC 2 Type 2, and GDPR compliance for Inworld TTS. Enterprise customers in healthcare, finance, and other regulated industries can now ensure that no customer text inputs or audio outputs are persistently stored in our systems after processing completes. How zero-data retention works: When enabled, all customer text inputs sent to our text-to-speech engine are processed in memory to generate audio. Once complete, both the text and the generated audio are immediately redacted from our logging systems. This applies to all text sent for speech synthesis, including requests using cloned voices. Enterprise-level security: Zero-data retention is configured at the workspace level, allowing customers with strict data privacy requirements to maintain compliance while still benefiting from the world's leading TTS model. Whether you're handling protected health information or sensitive financial data, you get the security guarantees needed for production deployment at scale. For questions about enterprise security features and HIPAA compliance, reach out at: https://lnkd.in/gCqmmzPa

Inworld TTS is now available on GMI Cloud! Our #1 ranked realtime voice AI models coupled with GMI’s high-performance, scalable GPUs give developers everything they need to build ultra-low latency and maximum efficiency realtime AI applications. The first 50 users can get $10 in Inworld TTS credits on GMI Cloud! Visit GMI on X for instructions on how to claim → https://lnkd.in/gETyYF3M Try Inworld TTS 1.5 on GMI Cloud today → https://lnkd.in/g3ymuGa8