Transparency about where the platform is going. Priorities shift with customer feedback — share what you want next and we will fold it into the next quarter.
Finishing the move to a globally distributed edge runtime — sub-second cold starts in every region, no operator pinning required.
Multi-user demo environment with shareable session links so you can test cross-language chat & video without writing a line of integration code.
React, React Native, Swift, Kotlin, Flutter, and JavaScript SDKs sharing a single API key — feature parity across web, mobile, and native.
Controls already in place across access management, change control, and incident response. Audit observation period now in motion.
Mint per-end-user JWTs from your backend so your mobile and web apps never need to ship a HyperBabel API key. Session-token pattern, 1h access + 30d refresh, server-side revocation.
1:1 and group voice rooms over a globally distributed SFU — opus codec, automatic gain control, and AI-interpreted transcripts on demand.
One-toggle session recording for video and live, with searchable multi-language transcripts indexed at the message level.
Opt-in E2EE for 1:1 and group rooms with zero-knowledge key exchange — designed for regulated chat (legal, health, finance).
First-party SDKs alongside the existing six. Same API key, same shape, idiomatic to each ecosystem.
Programmatic access to message volume, channel retention, translation usage, and per-key engagement — query directly or stream into your data warehouse.
Fine-tune AI translation on your domain’s terminology — preserve brand voice, handle proprietary jargon, opt out of public model training entirely.
Context-aware content filtering with policy customization per channel — toxicity, spam, sensitive content, and locale-specific norms.
Multi-region channel synchronization with conflict-free message ordering — write to the closest edge, replicate worldwide.
Single-tenant deployments and private-VPC options for regulated industries — same API surface, customer-controlled data residency.
Live speech recognition + real-time interpretation across 60+ languages in video calls (and later live streaming) — each participant hears every speaker in their own language, preserving prosody, identity, and cadence.
Disclaimer. This page expresses current intent. It is not a contractual commitment, an SLA, or a sales promise. Timing, scope, and priorities may shift based on customer feedback, security review, and operational readiness. Make purchasing or compliance decisions only against signed agreements.
This roadmap is co-authored with our customers. Tell us what should land next quarter.