This document explains the technical parameters and scoring for English extended multiple-choice speech assessments. It supports multi-option pronunciation evaluation and provides precise scoring.-Smart Science and Education Content Platform(SSECP)-阿里云帮助中心

This document provides technical details to help developers integrate the English extended multiple-choice speech assessment feature into their products.

Applicable scenarios:

You can set multiple answer options. A score is awarded only if the correct option is spoken with complete and accurate pronunciation.
Users can provide pronunciations that are not exact matches to the predefined answers. The engine identifies the closest matching answer and scores the pronunciation accordingly.
You can define incorrect answers. If a user provides one of these incorrect answers, it is marked as wrong.
If a user says multiple words or phrases that could be answers, the last one spoken is considered the final answer. For example, if a user says "key unkey", the final answer is "unkey".

Parameter description

Field	Type	Optional	Description
coreType	String	Required	en.pche.score indicates an English extended multiple-choice question.
rank	int	Optional	The scoring system for the total score.
precision	double	Optional	Sets the scoring precision. Supported values are 0.1, 0.5, and 1. The default value is 1. If you set a value other than 0.1 or 0.5, the system processes it as 1.
pronScale	int	Optional	1: The total score includes the pronunciation score. The formula is 50% + (50% × pronunciation score). 0: Default. The total score can only be the full score or 0. This field only accepts 0 or 1.
lm	array	Required	An array of dictionaries that sets the options for the extended multiple-choice question. (Note: The lm field is the lowercase form of LM.) Each text subfield represents a way to express a correct or incorrect answer. The keywords for correct or incorrect answers are enclosed in parentheses and separated by the "\|" symbol. The sentence outside the parentheses is combined with each keyword inside the parentheses to form a complete statement. For example: She is from (Tokyo\|Beijing\|London). This is internally split into the following three sentences: She is from Tokyo. She is from Beijing. She is from London.
key	array	Required	An array of dictionaries that sets the correct answers. Multiple correct answers are supported.
unkey	array	Required	An array of dictionaries that sets the incorrect answers.
phones	object	Optional	Specifies the pronunciation of a word. If a word has multiple pronunciations, separate them with the "\|" symbol. Use a space between each phoneme and between phonemes and the syllable separator "-". Example: "phones":{"directory":"d ih - 'r eh k - tr ih\|d ax - 'r eh k - t ax - r ih"};
attachAudioUrl	int	Optional	Specifies whether to return the audio URL in the assessment result. 0: Does not return the URL. 1: Returns the URL.

Parameter example

{
    "coreType": "en.pche.score",
    "rank": 2,
    "precision": 0.5,
    "attachAudioUrl":1,
    "lm": [
        {"text": "(Tokyo|Beijing|London)." },
        {"text": "Linda is from (Tokyo|Beijing|London)." },
        {"text": "Linda comes from (Tokyo|Beijing|London)." },
        {"text": "She is from (Tokyo|Beijing|London)." },
        {"text": "She comes from (Tokyo|Beijing|London)." }
    ],
    "key": [{ "text": "Tokyo" }],
    "unkey": [
        {"text": "Beijing"},
        {"text": "London"}
   ]
}

Sample response

Response description

Key	Type	Description	Recommendation
recordId	string	The unique ID of the audio file.	For reference.
tokenId	string	The unique ID of the user request.	For reference.
applicationId	string	The authorized assessment account, which is the AppKey.	Internal parameter.
audioUrl	string	The URL of the user's audio. By default, it is stored in the cloud for one month. Save it to your product server.	For reference.
dtLastResponse	string	The timestamp of the cloud response.	Internal parameter.
eof	int	0: Indicates that the response is not complete and more results will follow. 1: Indicates that all results for this assessment have been returned.	Internal parameter.
params	object	Echoes the assessment request parameters sent by the client.
- app	object
- - applicationId	string	The authorized assessment account, which is the AppKey.
- - userId	string	The user ID. A custom string for easier troubleshooting.
- - timestamp	string	The start time.
- - sig	string	The verification code.
- - clientId	string
- request	object	The node for scoring request parameters.
- audio	object	The node for audio parameters in the scoring request.

result	object	The assessment result.
- res	string	The name of the resource used for assessing this question type.	Internal parameter.
- version	string	The engine version, including the release date.	Internal parameter.
- rank	int	The scoring system.	Internal parameter.
- precision	float	The scoring precision.	Internal parameter.
- pretime	int	The time taken by the cloud to call the Start API, in ms.	Internal parameter.
- systime	int	The total time taken for the entire assessment process, in ms.	Internal parameter.
- delaytime	int	The time from when the audio feed ends to when the result is obtained on the cloud, in ms.	Internal parameter.
- wavetime	int	The duration of the audio, in ms.	For reference.
- overall	float	The total score.	Recommended field.
- fake_words	object		(This field is returned only if fake_words exist in the assessment text. Each word has a separate entry.)
- -char	string	A word that does not exist in the dictionary. The pronunciation is predicted or forged.
- info	object
- - snr	float	The signal-to-noise ratio (SNR). A higher value indicates clearer audio. Range: 0 dB to 40 dB. If this parameter affects the score, a corresponding tipId value is set.	For reference.
- - clip	float	Audio clipping occurred because the volume was too high. Range: 0 to 1. If this parameter affects the score, a corresponding tipId value is set.	Internal parameter.
- - volume	int	The recording volume. Range: 0 dB to 180 dB.	For reference.
- - tipId	int	Audio quality. For more information, see tipId message descriptions.	Recommended field.

Suggestions for displaying assessment results

Dimension	Description	Details
overall	Total score	The assessment score for the question type.

Scoring failure

If the result structure in the response is empty, the assessment has failed. You can retrieve the errId and error from the response and handle the error according to the error code.

Error codes

For more information about error codes, see Assessment error codes.