evidoria

← Back to browse

Good practice Imported

SABAY-Salita — ECAIR's AI Speech and Dyslexia Screening Pilot for Philippine DepEd Schools

Philippines · Manila · See the Philippines profile · See the Manila profile

Evidence: Descriptive / self-reported Top 81% 33/100 · Ask Evidence Copilot about this practice

DepEd's ECAIR piloted SABAY-Salita, an AI system that screens children's speech recordings for dyslexia and speech-sound disorders, in two National Capital Region schools from late 2025, aiming to catch cases manual screening misses and ease teachers' referral workload.

2 schools
Pilot schools (National Capital Region) (2nd semester 2025)

Details

Maturity
Pilot
Promoter
DepEd — Education Center for AI Research (ECAIR)
Period
2025–2026 (pilot rollout, two schools)
Keywords
special education, disability screening, speech-language pathology

Context

The Philippines' Department of Education (DepEd), through its Education Center for AI Research (ECAIR), developed SABAY-Salita to screen children's speech recordings for signs of dyslexia and speech-sound disorders.

Objectives

The tool is designed as a first line of defence, filtering learners who likely need referral to a specialist from those who do not, to catch disorders that often go undetected after preliminary teacher-led checks.

Activities

ECAIR built the system jointly with DepEd's Bureau of Learning Delivery - Student Inclusion Division, alongside education experts, healthcare professionals and special-education practitioners, and began pilot testing in two schools in the National Capital Region in the second semester of 2025.

Results

DepEd publicly presented SABAY-Salita in September 2025 as part of a wider set of AI projects alongside its national push on AI transparency and integrity, and ECAIR has stated an intention to expand the pilot into 2026 with more regions, larger datasets and additional training for teachers and learners.

Conclusions

As of the pilot stage, no independent accuracy figures, false-positive/negative rates or downstream intervention outcomes have been published; the documented evidence covers the tool's design rationale and initial two-school deployment, not its screening performance.

Implementation
Indicative cost
Medium (€50k–€500k)
Time to results
Medium (1–3 years)
Staffing & skills
DepEd Education Center for AI Research (ECAIR) technical team, DepEd Bureau of Learning Delivery - Student Inclusion Division, Healthcare professionals and special-education practitioners co-developing the tool

Conditions for success

  • Inter-agency co-development spanning education, health and special-education expertise
  • Phased rollout starting with a small, controlled two-school pilot before wider expansion
  • Planned additional training for teachers and learners as the pilot expands

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful