General Speaking Placement Test

Introduction

This document describes the Speaking Placement Test developed using the Speech Assessment API (SAAPI) from Language Confidence. Language Confidence is a global provider of AI-powered language assessment technologies that deliver consistent and scalable results for English language proficiency testing. This test evaluates English speaking proficiency across key CEFR levels (A1 to C2), using real-world tasks that assess pronunciation, fluency, grammar, vocabulary, and content relevance. The design integrates Language Confidence’s SAAPI automated scoring capabilities to ensure consistent, objective, and reliable placement outcomes.

Test Structure

The test consists of 15 questions, arranged progressively from basic to advanced language tasks. The early section focuses on fundamental communication skills and familiar everyday situations. The middle section introduces intermediate-level interaction and reasoning tasks. The final section explores more advanced language use, encouraging extended responses and discussion of broader or more abstract topics. This progression allows test takers to demonstrate their English speaking ability across a wide range of CEFR levels while remaining age-appropriate and engaging for adults and young adults.

Scoring Rubric

Each question is allocated maximum points based on linguistic complexity. Test takers may earn partial credit depending on the quality and completeness of their answers. Lower-level tasks carry up to 5 points, mid-level tasks up to 7 points, and upper-level tasks up to 8 points, for a total of 100 maximum points.

Points Allocation:

- A1–A2 (Questions 1–5): 5 points each

- B1–B2 (Questions 6–10): 7 points each

- C1–C2 (Questions 11–15): 8 points each

Total possible score: 100 points

CEFR Score Ranges

Overall test scores map to CEFR proficiency levels as follows:

- A1: 0–15

- A1+: 16–25

- A2: 26–40

- B1: 41–55

- B1+: 56–65

- B2: 66–80

- C1: 81–90

- C2: 91–100 Scoring Methodology

SAAPI evaluates each spoken response across Pronunciation, Fluency, Grammar, Vocabulary, and Content Relevance. Scores from these dimensions are aggregated to yield the per-question point value specified in the rubric. Partial credit is awarded where appropriate to reflect partially correct but meaningful language use.

Reliability of Test Results

The Speaking Placement Test benefits from the proven reliability of the Speech Assessment API (SAAPI) provided by Language Confidence. SAAPI applies advanced speech recognition and natural language processing technologies to ensure consistent, objective, and reproducible scoring. By analyzing pronunciation, fluency, grammar, vocabulary, and content relevance with standardized algorithms, the test minimizes human bias and variability. Language Confidence positions SAAPI as a robust and scalable solution for assessing spoken English, which supports the reliability of the placement results. While no automated test is entirely free of external factors (e.g., recording quality, background noise), SAAPI’s detailed phonemic analysis and comprehensive scoring methodology provide a dependable measure of English speaking proficiency across CEFR levels.

Last updated