Sanas

This Thanksgiving, we’re thankful for our incredible team, our customers, and the community who brings our mission to life. If you're celebrating across distance or languages, the Sanas Language Translation mobile app can help everyone join the conversation. Get it here: https://lnkd.in/gck6MdF9 Wishing everyone a holiday filled with connection, reflection, and gratitude.

Thank you Analytics India Magazine (AIM) for spotlighting this thoughtful conversation with Nitesh Varshney, our SVP of Engineering! The lively discussion spans everything from the moment of injustice that sparked Sanas’ origin story and the science behind our real-time speech AI to the real-world impact it’s having on agents, enterprises, and global communication. If you haven’t watched yet, take a moment to check out the clip!

Voice Activity Detection is mission-critical to modern speech technology systems, but most real-time VAD systems weren't designed for real-world conditions: background noise, far-field mics, and overlapping voices. And when VAD fails, everything downstream becomes less reliable. So we redesigned VAD from the ground up. Sanas VAD was built specifically for real-time, low-latency streaming applications and challenging acoustic environments. Sanas Machine Learning Researcher Álvaro Escudero Barrero walks through the full technical deep dive, from architecture and benchmarks to audio examples. Take a look: https://bit.ly/49G1LwP Sharath Keshava Narayana, Shawn Zhang, Anant S., Monal P., Saralynn Malott, MS., Michael L. Johnson, Nitesh Varshney, Ashish Nagpal, Maxim Serebryakov, Marty Massih Sarim #soundslikeinnovation #soundslikesanas

We’re thrilled to share that Sanas Language Translation has been named “Most Innovative App” in the Expo App Awards out of 1200 contenders! We started with a simple idea: what if two people could speak different languages and still hear each other naturally, in real time, without losing tone, emotion, or identity? This app came to life the way the best projects do — a tight team, a big vision, and the right tools to move quickly. Expo gave us the freedom to stay focused on what mattered most: low-latency translation, voice identity, and a truly natural conversation experience. As Jon Samp, Head of Product at Expo, shared during the award ceremony: “My wife's entire family speaks Italian, and Sanas takes my English and turns it into fluent Italian almost instantly. I haven't seen it done before in a way that is convincing. It inspires imagination: an example of building something super useful that gets out of your way” Huge thanks to Expo for the recognition, and to the Sanas team whose creativity and dedication made this breakthrough possible! Check out why Expo chose Sanas at https://expo.dev/awards Sharath Keshava Narayana, Shawn Zhang, Anant S., Monal P., Saralynn Malott, MS., Michael L. Johnson, Nitesh Varshney, Ashish Nagpal, Maxim Serebryakov, Marty Massih Sarim, Jason Lin, Scott Hickmann, Sushen K., John Hong, Vivekanand Balakrishnan

Grateful for our incredible Philippines team, whose energy, warmth, and teamwork make such a meaningful impact every day! Special shoutout to our amazing demoers, who showcase Sanas at its best in every conversation. Catch a glimpse of the fun from our Philippines Holiday Party — and if you'd fit right in, we’re growing! https://bit.ly/3KhDpiA

How do you teach an AI system to understand who is speaking, where they are, and what to preserve — all in real time? In this technical deep-dive, Senior Research Scientist Anatole Chivot takes you inside the science behind Speech Enhancement 1.0, including: 🔀 Dual-decoder architecture modeled after human auditory processing 📡 Physics-based acoustic simulation for distance-aware voice isolation 🎧 Real-world evaluation across diverse and adversarial acoustic conditions For engineers, researchers, and anyone building at the edge of speech technology, this is a peek behind the scenes. Explore the science behind Speech Enhancement 1.0: https://lnkd.in/gwAdd83M Sharath Keshava Narayana, Shawn Zhang, Anant S., Monal P., Saralynn Malott, MS., Michael L. Johnson, Nitesh Varshney, Ashish Nagpal, Maxim Serebryakov, Marty Massih Sarim

What happens when you stop following the metro playbook and start building opportunity where people actually live? In his new article, Ashish Bisaria shares how a rural-first strategy paired with Sanas boosted Trajector's hiring efficiency from 10% to 33% and delivered one of the strongest retention rates in the industry: 0.8% monthly attrition. The result: a workforce that stays, grows, and leads with dignity. Read the behind-the-scenes look at scaling with purpose: https://bit.ly/3LGayVI Sharath Keshava Narayana, Shawn Zhang, Anant S., Monal P., Saralynn Malott, MS., Michael L. Johnson, Nitesh Varshney, Ashish Nagpal, Maxim Serebryakov, Marty Massih Sarim

🚀 Introducing Sanas Speech Enhancement 1.0! Speech Enhancement combines a new proprietary AI model trained on massive real-world datasets to deliver speech that preserves the fullness of emotion, tone, and identity even in the most chaotic environments. ✅ Best-in-Class Robustness: Performs consistently across unpredictable, multi-speaker, and high-noise environments. ✅ Standard & Voice Isolation Modes: Standard Mode preserves all nearby speakers, and Voice Isolation Mode enhances on only the foreground speaker for maximum clarity and privacy. ✅ Ultra-Fidelity Audio: Now supporting 24 kHz for the full warmth, texture, and detail of the human voice. ✅ Smart Ringtone Passthrough: Configurable ringtone passthrough keeps agents and professionals alert for incoming calls. Experience Speech Enhancement 1.0 and hear the difference: https://lnkd.in/g2GTeBjf Sharath Keshava Narayana, Shawn Zhang, Anant S., Monal P., Saralynn Malott, MS., Michael L. Johnson, Nitesh Varshney, Ashish Nagpal, Maxim Serebryakov, Marty Massih Sarim

Building a universal translator has been a dream for decades — from Star Trek to Silicon Valley. Jason Lin and Scott Hickmann share how their Sanas Language Translation team tackled that challenge head-on, combining clause-aware LLMs, real-time speech AI, and voice-preserving translation to make conversations effortless across languages. Read how we solved the “translation trinity” of accuracy, speed, and naturalness: https://bit.ly/4ovP646 Sound interesting? We’re hiring! Come build the world’s most natural real-time translator with us: https://bit.ly/4i7bgav Sharath Keshava Narayana, Shawn Zhang, Anant S., Monal P., Saralynn Malott, MS, Michael L. Johnson, Nitesh Varshney, Ashish Nagpal, Maxim Serebryakov, Marty Massih Sarim

AI is reshaping customer experience, but as CX Today highlights, success depends on understanding the human voice. Sanas CEO Sharath Keshava Narayana spoke with Rhys Fisher about why so many AI initiatives stall and how speech AI that enables clarity and understanding is more important now than ever. Read the full feature: https://lnkd.in/eVKqmRbi #CustomerExperience #ConversationalAI #VoiceAI #CXInnovation #soundslikeclarity #soundslikesanas