My Speaking Score (MSS), a leader in automated speech evaluation technology, is proud to announce the release of its new API, designed to streamline the assessment of TOEFL®-style speaking responses. This powerful tool is tailored for developers, educational technology (EdTech) companies, and institutional partners, providing a fast, scalable, and automated solution for speaking assessment.
The MSS API offers an integrated approach to scoring speaking responses by analyzing audio files and delivering precise, SpeechRater™-aligned scores. The API evaluates a wide range of speech features, offering overall predictions and detailed feedback that can be incorporated into external platforms and tools for a variety of applications.
A Comprehensive Solution for Speaking Assessment
The API is designed to cater to various users in the education technology space, particularly those in TOEFL preparation, digital classrooms, and automated language testing. It provides a comprehensive solution by scoring speaking responses, offering sub-scores across several dimensions including fluency, pronunciation, grammar, and vocabulary. It also returns granular feature values such as speaking rate, pause frequency, and lexical density, among others.
The detailed feedback from the MSS API is valuable not only for scoring but also for helping users improve their speaking abilities. Each response includes optional percentile ranks based on MSS’s internal benchmarked data set, giving users context for their performance. Additionally, for broader applicability, some endpoints estimate Common European Framework of Reference (CEFR) levels and TOEFL speaking band equivalents, which is particularly useful for tools aligned with other proficiency exams like IELTS or DET.
Key Features and Capabilities
The MSS API offers a robust range of capabilities that provide ease of integration and flexibility for various use cases:
- Scoring of Speaking Responses: The API accepts audio files in formats such as .wav and .mp3 and returns scores on a 0.0–4.0 scale, including sub-scores for fluency, pronunciation, grammar, and vocabulary.
- Detailed Feedback: The API provides detailed feature values that help evaluate specific aspects of speaking performance, such as speaking rate and lexical diversity.
- TOEFL and CEFR Estimates: It also estimates TOEFL speaking bands and CEFR levels to support broader use cases, ensuring compatibility with multiple English proficiency tests.
- Easy Integration: The API is designed for integration into various platforms, enabling TOEFL prep companies, EdTech tools, and academic institutions to streamline speaking assessments and feedback.
Secure and Scalable Authentication
The MSS API employs secure authentication methods, utilizing API keys for each partner. This ensures that every partner receives a unique key, allowing them to manage usage quotas efficiently and securely. Additionally, the API supports high-volume integrations, such as batch processing, making it an ideal solution for institutions or companies with large-scale requirements.
Supported Input and Flexible Usage
The API accepts audio files in .wav or .mp3 formats, with a maximum length of 60 seconds. It can also accept JSON payloads containing metadata, including task type and language, with optional fields for user identification for tracking purposes.
Practical Use Cases Across the Education Sector
The MSS API is designed to serve a variety of users, including:
- TOEFL Prep Platforms: Educational companies and platforms can use the API to automatically score student responses and provide real-time feedback within their applications.
- EdTech SaaS Tools: Digital classrooms and coaching platforms can integrate the API to incorporate automated speaking assessments into their offerings, helping learners track progress and improve their speaking skills.
- Universities and Schools: Educational institutions can automate oral placement tests and monitor speaking improvement over time, making the evaluation process more efficient and standardized.
Example Use of the API
To illustrate the ease of integration, the MSS API offers a simplified example endpoint for scoring speaking responses. Partners can make a POST request to the /v1/speaking/score endpoint, providing an audio file URL along with relevant metadata like task type and language. The response will include an overall score, sub-scores for various features, percentile ranks, and estimates of TOEFL and CEFR equivalents.
Flexible Access and Integration Options
Access to the MSS API is granted through API partnership agreements, providing partners with the necessary resources to integrate the API into their existing systems. Some partners may also be offered access to raw feature outputs, enabling advanced analytics and custom reporting.
The MSS API supports batch processing, making it an excellent tool for organizations that need to process large volumes of speaking assessments quickly and efficiently. This is especially valuable for platforms that need to handle multiple users at once, ensuring seamless integration and scalability.
About My Speaking Score
My Speaking Score (MSS) is a leader in speech evaluation technology, providing reliable and scalable tools for assessing spoken language proficiency. With a focus on accuracy, speed, and ease of use, MSS delivers automated, AI-driven scoring solutions that meet the needs of educational institutions, exam preparation platforms, and EdTech companies worldwide.
To learn more about My Speaking Score’s API and its integration capabilities, visit My Speaking Score API.
