ELSA

ELSA (English Language Speech Assistant) builds AI speech-recognition and pronunciation-assessment technology for non-native English speakers, powering the ELSA Speak consumer app. The ELSA API is a partner/B2B speech-assessment service that scores recorded or streamed English audio across pronunciation, fluency, intonation, grammar, and vocabulary, returning sentence, word, and phoneme-level feedback.

3 APIs 0 Features

Speech AssessmentPronunciationSpeech RecognitionLanguage LearningAI

APIs

ELSA Pronunciation Assessment API (Scripted)

Scripted speech scoring. The caller submits an audio recording together with the expected script/sentence and receives sentence, word, and phoneme-level pronunciation scores, mi...

ELSA Realtime Streaming Speech API

Realtime, packet-by-packet audio scoring over WebSocket. Scripted streaming connects to wss://api.elsanow.io/api/v2/ws/score_audio and unscripted (open-ended) streaming connects...

ELSA Unscripted Speech Assessment API

Open-ended (unscripted) speech assessment. The caller submits spontaneous English audio with no reference script and receives a spoken transcript plus pronunciation, prosody/int...

Collections

ELSA Speech Assessment API

OPEN

Pricing Plans

Elsa Plans Pricing

3 plans

PLANS

Rate Limits

Elsa Rate Limits

4 limits

RATE LIMITS

FinOps

Elsa Finops

FINOPS

Resources

👥

GitHubOrganization

Sources

opencollection: 1.0.0
info:
  name: ELSA Speech Assessment API
  version: '1.0'
request:
  auth:
    type: apikey
    key: Authorization
    value: ELSA {{apiToken}}
    in: header
items:
- info:
    name: Scripted
    type: folder
  items:
  - info:
      name: Score a recorded audio file against an expected script (batch).
      type: http
    http:
      method: POST
      url: https://api.elsanow.io/api/v2/score_audio
      body:
        type: multipart-form
        data: []
    docs: Single-file (batch) scripted scoring. Submit a recorded audio file plus the expected script and an api_plan; returns
      sentence, word, and phoneme-level pronunciation scores with fluency and intonation. Request/response schemas are partner-gated.
- info:
    name: Unscripted
    type: folder
  items:
  - info:
      name: Score a recorded audio file with no reference script (batch).
      type: http
    http:
      method: POST
      url: https://api.elsanow.io/api/v1/score_audio_plus
      body:
        type: multipart-form
        data: []
    docs: Single-file (batch) unscripted scoring. Submit spontaneous audio and an api_plan; returns a transcript plus pronunciation,
      prosody, fluency, grammar, and vocabulary analysis. Request/response schemas are partner-gated.
bundled: true