Kid Smart AI Fluency Assessment API offers compatible results with SoapBox Labs API output.
Change one line of code from https://api.soapboxlabs.com/v1/async/request/speech/fluency
to
https://api.kidsmart.ai/v1/audio/fluency
Then get results that are SoapBox compatible:
See our playground.kidsmart.ai for a demo of our technology if you have reading assessments on hand. We also provide assessment examples in different environments.
Identical Inputs:
| Description | Kid Smart AI Fluency Inputs | SoapBox Labs Fluency Inputs |
|---|---|---|
| The audio file to be analyzed | file | file |
| The reference text against which the speech contained in the audio file should be analyzed | reference_text | reference_text |
| A unique ID that represents the speaker in the audio file. | user_token | user_token |
| The alphanumeric ID of the model. | model_id | model_id |
Identical & Improved Outputs:
| Description | Kid Smart AI Fluency Outputs | SoapBox Labs Fluency Outputs |
|---|---|---|
| The user_id specified in the request | user_id | user_id |
| The language contained in the audio file being analyzed | language_code | language_code |
| A unique identifier for the request | result_id | result_id |
| The UTC time the request was processed at | time | time |
| The duration of the audio file in seconds. At Kid Smart AI, this is the number of seconds you are charged for. | audio_duration | audio_duration |
| Timestamp of first word of the assessment | assessStartTime | N/A |
| Timestamp of the last word of the assessment | assessEndTime | N/A |
| The Fluency results object returned (see below) | results | results |
| The number of differences between the reference and transcription text. | results.num_differences | results.num_differences |
| The number of times a word has been substituted for another. | results.substitution_count | results.substitution_count |
| The number of times a word has been inserted. | results.insertion_count | results.insertion_count |
| The number of times the child correctly said a word from the reference text. | results.correct_count | results.correct_count |
| The number of times a word from the reference text was not said. | results.deletion_count | results.deletion_count |
| The text that is expected to be read. | N/A | results.reference_text |
| A breakdown of the individual words. | results.text_score | results.text_score |
| The computed transcription from the audio file. | N/A | results.transcription |
| The number of words in the reference_text. | results.word_count | results.word_count |
| The number of times the child repeated a word. | results.repetition_count | results.repetition_count |
| Optional input for SoapBox Labs API. Not necessary for Kid Smart AI | N/A | results.last_word |
| Number of correct words read per minute (words per minute) | results.WPM | N/A |
| Number of words read correctly divided by total number of words, times 100 (percentage) | results.accuracy_score | N/A |
See our API Docs for more information: Kid Smart AI Fluency API Docs