Skip to main content

Single-word/sentence assessment usage – Upload via Data Stream​

Single-word/single-file assessment for a single file (batch) mode. The developer’s server/app sends a TCP call to the ELSA server with all the necessary information for the assessment. The ELSA server processes the request synchronously and returns the result in the same call once it’s ready.

To perform an assessment on a word or sentence, the developer must provide:

•   The word/sentence expected to be found in the audio
•   The audio data (sent as a Base64-encoded string)
•   The API token provided by ELSA
Header Parameters
    Authorization: Elsa string required
Request Body

Audio data must be provided as a Base64-encoded string:

- audio_data: <base64_data_string> → The raw audio data encoded as a Base64 string. This method allows sending audio directly in the request body without requiring a file upload or external URL reference.
    sentence string required
    audio_data byte required

    Possible values: Value must match regular expression <base64_data_string>

    Base64-encoded audio data string.

Responses

OK

Schema
    api_version string
    total_time number
    utterance object[]
  • Array [
    • initial_silence boolean
    sentence string
    sentence_id integer
    total_time number
    attempt_type string
    snr number
    decision string
    ipa string
    intonation_score number
    fluency_score integer
    speed_metrics object
    words_per_minute number
    syllables_per_minute number
    phones_per_minute number
    articulated_words_per_minute number
    articulated_syllables_per_minute number
    articulated_phones_per_minute number
    pronunciation_rate_metrics object
    correct_words_per_minute number
    correct_syllables_per_minute number
    correct_phones_per_minute number
    correct_words_ratio number
    correct_syllables_ratio number
    correct_phones_ratio number
    pausing_metrics object
    pause_ratio number
    pause_ratio_without_initial_sil number
    pausing_score_percentage integer
    advanced_metrics_confidence string
    ielts_score number
    cefr_score string
    toefl_score integer
    toeic_score integer
    pte_score integer
    pronunciation_score number
    pronunciation_score_partial number
    eps_score number
    recording_quality string
    words object[]
  • Array [
    • start_index integer
    end_index integer
    start_time number
    end_time number
    trans_arpabet string
    decoded boolean
    ipa string
    nativeness_score integer
    decision string
    word string
    text string
    phonemes object[]
  • Array [
    • start_index integer
    end_index integer
    text string
    trans_arpabet string
    start_time number
    end_time integernumber
    decision string
    nativeness_score integer
    phoneme_error string
    phoneme_error_arpabet string
    ipa string
  • ]
    • word_stress object[]
  • Array [
    • start_index integer
    end_index integer
    decision string
    stress_level_measured string
  • ]
  • ]
  • ]
    • success boolean
Loading...