Speak
Onboarding assessment builds an initial learner profile and generates a personalized curriculum; AI supports open-ended conversational practice with real-time speech recognition. No SRS or long-term vocabulary tracking — saved phrases go to a passive Phrasebook, not systematically reintegrated. Per-session context is strong; cross-session memory is structurally weak.
At a glance
- Type
- Onboarding curriculum + per-session conversation
- Tier
- T2
- Created
- 2016 (Speak founded by Connor Zwick and Andrew Hsu; Thiel Fellowship; San Francisco; YC backed)
- Latest release
- not applicable — not OSS
- License
- not applicable — not OSS
- GitHub
- not applicable — no GitHub repo
- Pricing
- Free trial; Premium plans (multiple tiers); Enterprise (SSO/white-label)
- Funding
- $162M total raised; Series C $78M at $1B valuation (Accel/OpenAI/Khosla/YC) Dec 2024
Taxonomy
- storage
- relational
- retrieval
- injection
- persistence
- cross-session
- update
- overwrite
- unit
- profile
- governance
- inspectable
- conflict
- overwrite
When to use
Optimised for: learner profile + spaced repetition + assessment-driven adaptation
Anti-fit: not for non-educational use cases
Pros & cons
Pros
Strongest real-time conversational AI of any language app — speech recognition, scenario roleplay, instant feedback are tightly integrated; OpenAI-backed.
Cons
Cross-session memory acknowledged as a gap by reviewers — no SRS, no systematic error tracking across conversations; Phrasebook is passive rather than adaptive.
Claims & capabilities
10M+ users; doubling annually for five years. $78M Series C at $1B (Dec 2024); $100M revenue 2025.
Technical surface
- API surface
- searched not found
- Backend storage
- searched not found
- Deployment
- iOS + Android + Web + Enterprise embedded
- Embedding model
- searched not found
- Multi-tenancy
- searched not found
- MCP
- no MCP support advertised — vertical product, no MCP server / client integration documented
- A2A
- no A2A protocol support advertised — vertical product, no A2A integration documented
- OpenTelemetry
- no OpenTelemetry integration advertised — vendor logs/observability not publicly documented
Similar systems
Other vertical / domain-specific ai memory in the catalog, ranked by inbound references.
- NVIDIA ReMEmbR T3
Builds long-horizon memory by captioning video segments with VILA, storing captions with timestamps + 3D position coordinates in MilvusDB. At query time, LLM iterates retrieval across text, time, and position modalities. Deployed on Nova Carter robot (Jetson Orin).
- Abridge T1
Clinician-assist ambient documentation. Source mapping: every AI-generated summary element traced back to the source utterance. Audit-and-trust layer over episodic memory. Built on proprietary 1.5M+ medical-encounter dataset.
- ASAPP GenerativeAgent T1
Treats memory as first-class architecture. Captures the digital footprint of every interaction; retrieves preference and history at engagement time. Public example: airline knowing a frequent flyer wants aisle seats with her son — preference-aware, not just history-lookup.
- BenevolentAI T1
Target identification / drug repurposing / mechanism tracing. 85+ data sources, petabyte-scale, rebuilt every few weeks. Wet-lab results re-enter the graph and shift downstream predictions — institutional experimental memory closing a feedback loop.
- Causaly T1
Drug discovery / target identification / causal mechanism tracing. The graph is the memory: 7 years of curated biomedical cause-effect relationships compounding with each new ingestion. Scientific RAG retrieves from a versioned causal substrate.
- Character.ai T1
Chat Memories (user-defined facts), auto-memories for c.ai+ subscribers, pinned memories, in-context retention. PipSqueak 2 model (April 2026) reduces in-conversation drift. Memory Visualization meter forthcoming.
Related systems
References (1)
- OpenAI GPT family (GPT-5 / GPT-4o / o3 / o4) depends on at runtime — tightly integrated; OpenAI-backed.