ARASH
Sign inOpen dashboard
Voice waveform transforming into layered data streams
Adaptive Responsive AI for Speaking Hub

Every spoken word becomes measurable progress

An enterprise English speaking platform for junior high school. Students speak, eleven AI engines analyse, and students, teachers, parents, researchers and administrators each get the view they need — with evidence they can play back.

Enter the platform Explore modules
6
skill dimensions
11
AI engines
<20s
feedback turnaround
5
user roles

Assessment

Six dimensions, one composite band

Each score links back to the exact moment in the recording that produced it, and every teacher override is tracked as a revision.

Pronunciation

78

Grammar

71

Vocabulary

83

Fluency

66

Comprehension

88

Pragmatic

74

AI services layer

From spoken utterance to next activity

A student speaks once. The pipeline transcribes, analyses six dimensions, composes audience-specific feedback, updates the learner model, and refreshes analytics.

  1. 0

    Student speaks

    Resumable chunked upload

  2. 1

    Speech Recognition AI

    Transcript, word timings, confidence

  3. 2

    Conversation AI

    Scaffolded dialogue turns and hints

  4. 3

    Pronunciation AI

    Phoneme accuracy, stress, intonation

  5. 4

    Grammar AI

    Error taxonomy, severity, correction

  6. 5

    Vocabulary AI

    Range, CEFR band, target-word coverage

  7. 6

    Fluency AI

    Rate, pauses, disfluency, mean length of run

  8. 7

    Comprehension AI

    Relevance, accuracy, instruction following

  9. 8

    Pragmatic AI

    Register, politeness, turn-taking

  10. 9

    Feedback AI

    Student, teacher and parent feedback variants

  11. 10

    Adaptive Recommendation AI

    Mastery update and next-activity plan

  12. 11

    Learning Analytics AI

    Trajectories, cohorts, at-risk signals

  13. ✓

    Dashboard

    Realtime push to every role

System architecture

Six layers, strict boundaries

No layer calls upward, no module reads another module's tables, and no vendor SDK exists outside the AI gateway.

L1

Presentation Layer

Role-aware clients, speech capture, realtime rendering

Role workspaces
Speaking client
Realtime client
Design system
L2

Application Layer

Gateway, identity, orchestration, quotas, contracts

API gateway
Middleware pipeline
Use-case orchestrators
Realtime gateway
L3

AI Services Layer

AI gateway plus specialized engines and rubric aggregation

Model registry
Analysis engines
Rubric aggregator
Safety filters
L4

Business Logic Layer

Domain invariants, pedagogy policy, workflows, events

Domain modules
Policy engine
Workflow engine
Event publisher
L5

Data Layer

OLTP, OLAP, object storage, cache, research vault

Six data domains
Object storage
Cache cluster
Governance
L6

Infrastructure Layer

Cloud, edge, queues, secrets, observability, DR

Edge & network
Worker pools
Observability
Backup & DR

Backend architecture

Ten disciplined concerns

Each concern has one owner and one place to change.

Controller

Transport handlers, DTO mapping, response envelopes

Middleware

Trace, authN, tenancy, authZ, validation, rate limit

Service Layer

Use-case orchestration and transaction boundaries

Repository Layer

Aggregate persistence, unit of work, row-level tenancy

API

Versioned public, private, AI, admin, analytics, research surfaces

Authentication

Tokens, sessions, SSO, guardian consent gating

Caching

Edge, HTTP, distributed, AI response cache

Logging

Application, audit, security, AI interaction streams

Queue

Scoring, analytics, notify, export lanes with DLQ

Storage

Private audio buckets, signed URLs, lifecycle tiers

API architecture

Nine versioned surfaces

Separated by audience so security, quotas and deprecation are independent.

Authentication APISessions, tokens, SSO, step-up
User APIProfiles, roles, links, consent
Learning APICurriculum, classes, assignments, paths
Conversation APIDialogue sessions, voice, transcripts
Assessment APIAttempts, rubrics, scores, feedback
Recommendation APIMastery, plans, spaced repetition
Analytics APIStudent, teacher, usage, performance
Research APIStudies, cohorts, snapshots, exports
Admin APIConfig, prompts, quotas, audit

Roles

Five audiences, one source of truth

Access is deny-by-default: role, tenant, ownership, class membership and consent scope all have to agree.

Student

Speak, get scored, improve

Daily speaking plan, live AI conversation partner, evidence-linked feedback.

  • Practice & submit attempts
  • View own scores
  • Play back recordings

Teacher

See the class, intervene early

Class health, assignment delivery, score review and override, skill heatmaps.

  • Manage assignments
  • Override scores
  • View class analytics

Parent

Follow progress, control consent

Weekly digests, plain-language summaries, privacy and consent centre.

  • View child progress
  • Manage consent
  • Message teachers

Researcher

Study outcomes ethically

Cohort builder, instruments, de-identified dataset snapshots and exports.

  • Build cohorts
  • Administer instruments
  • Request exports

Administrator

Govern the platform

Tenancy, users and roles, AI prompt governance, cost and health, audit log.

  • Manage users & roles
  • Govern AI prompts
  • Read audit log

Modular

New modules register in; existing modules never change.

Governed

Six data domains with tenancy, retention and consent.

Safe for minors

Consent-gated AI, short voice retention, audit trail.

Production-ready

10k concurrent learners, 99.9% on the learning path.

Ready to see the workspace for your role?

Choose a role and open the dashboard. Every module, metric and panel maps to the ARASH architecture blueprint.

Continue to sign in