This document explains the technical parameters, feedback dimensions, and result analysis for English recitation evaluation. It supports developers in integrating the pronunciation assessment feature.-Smart Science and Education Content Platform(SSECP)-阿里云帮助中心

This document provides the technical specifications for the English recitation evaluation topic type, which developers can use for product feature integration.

Description

This feature evaluates a user's recitation by comparing their pronunciation to the reference text.
The feedback includes an overall score, a score for each word, a fluency score, an integrity score, and a URL for the user's audio.

Evaluation request parameters

Field	Type	Optional	Description
coreType	String	Required	en.recite.score indicates the English recitation evaluation type.
refText	String	Required	The text to be recited.
symbol	int	Optional	Specifies whether to include punctuation in the results. Set to 0 or 1. The default is 0, which means punctuation is not returned.
feedback	int	Optional	Specifies whether to return recitation content in real time. Set to 0 or 1. The default is 0, which indicates non-real-time mode.
attachAudioUrl	int	Optional	Specifies whether to return the audio URL in the evaluation result. 0: Does not return the URL. 1: Returns the URL.

Parameter example

{
    "coreType":"en.recite.score",
    "feedback":1,
    "refText":"Rain is falling all around, It falls on field and tree, It rains on the umbrella here, And on the ships at sea"
}

Sample response

{
    "request_id":"26efa3e64c0c11ecb9d8f348945006b1",
    "applicationId":"a148",
    "dtLastResponse":"2021-11-23 11:19:34:381",
    "connect":Object{...},
    "params":{
        "app":Object{...},
        "audio":Object{...},
        "request":{
            "request_id":"26efa3e64c0c11ecb9d8f348945006b1",
            "refText":"Rain is falling all around, It falls on field and tree, It rains on the umbrella here, And on the ships at sea",
            "coreType":"en.recite.score"
        }
    },
    "recordId":"11ec4c0c2768b6aa8576a1480691312f",
    "refText":"Rain is falling all around, It falls on field and tree, It rains on the umbrella here, And on the ships at sea",
    "eof":1,
    "result":{
        "overall":78,
        "wavetime":11700,
        "precision":1,
        "version":"0.0.80.2021.11.22.14:14:57",
        "systime":11756,
        "res":"eng.recite.mfcc.0.1",
        "details":[
            {
                "char":"Rain",
                "dur":420,
                "score":99,
                "end":660,
                "start":240
            },
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            {
                "dp_type":1,
                "char":"It",
                "dur":0,
                "score":0,
                "end":0,
                "start":0
            },
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...},
            Object{...}
        ],
        "delaytime":68,
        "info":{
            "tipId":0,
            "clip":0,
            "snr":10.773011,
            "volume":86
        },
        "integrity":87,
        "rank":100,
        "rhythm":0,
        "fluency":91,
        "pretime":1126,
        "pron":75
    },
    "cloud_platform":Object{...}
}

Response description

Key	Type	Description	Recommendations
tokenId	string	The unique identifier for the user request.	For reference.
applicationId	string	The authorized account, which is the user's AppKey.	Internal parameter.
audioUrl	string	The download URL for the user's audio file. Audio files are stored for about 20 days by default. To store them longer, download them to your own server. The domain name varies based on the access region. The returned URL is the one to use.
recordId	string	The unique identifier for the audio file.
refText	string	The text to be recited.
dtLastResponse	string	The timestamp of the cloud response.	Internal parameter.
eof	int		Internal parameter.
params	object	Echoes the evaluation request parameters sent by the client.
- app	object
- - applicationId	string	The authorized account.
- - userId	string	The user identifier for your product. This can be a custom string. Use a unique userId for each user to make troubleshooting easier.
- - timestamp	string	The start time.
- - sig	string	The verification code.
- request	object
- - coreType	string	The engine to use.
- - tokenId	string	The unique identifier for the user request.
- - refText	string	The text to be recited.
- - attachAudioUrl	int	Specifies whether the scoring result includes the audio URL.
- audio	object
- - sampleRate	int	The sample rate.
- - channel	int	The sound channel.
- - sampleBytes	int	The number of bytes per sample. Supports single-byte and double-byte.
- - audioType	string	The audio format.
result	object	The evaluation result.
- overall	float	The overall score.	Internal parameter.
- systime	int	The total time consumed by the entire evaluation process, in ms.
- res	string	The name of the resource used for this evaluation type.	Internal parameter.
-fluency	object	Fluency details.
-version	string	The engine version, including the release date.
- integrity	float	The integrity score.	Recommended field.
- pretime	int	The time consumed by the cloud to call the Start API, in ms.
- pron	float	The pronunciation score.	Internal parameter.
-delaytime	int	The time from when the cloud finishes feeding the audio to when the result is obtained, in ms.	Internal parameter.
-wavetime	int	The duration of the audio.	For reference.
- info	object
- - snr	float	The signal-to-noise ratio (SNR).	For reference.
- - clip	float	The audio volume is too high, causing clipping. The range is from 0 to 1. If this parameter affects the score, a corresponding tipId value is set.	Internal parameter.
- - tipId	int	Audio quality.	Recommended field.
-- volume	int	The recording volume. The range is from 0 to 180 dB.	For reference.
- details	array	Detailed scores for each word in the final result.	Recommended field.
- - score	float	The pronunciation score for the word.
- - char	string	Word
- - dp_type	int	Over-reads and missed reads	0: Normal pronunciation. This is usually not displayed, but is always displayed in real-time feedback mode. 1: The word was omitted. 2: The word was repeated.
- - end	int	The end time of a phoneme in milliseconds (ms).
- - start	int	The start time of the phoneme, in milliseconds (ms).

Response description

Dimension	Description	Details
overall	Overall score	Overall rating.
pron	Pronunciation score
fluency	Overall fluency score
integrity	Integrity score

Response for a failed evaluation

If the result structure in the response is empty, the evaluation failed.
The response contains an errId and an error message. You can use the error code to determine the next steps for your product.

Error codes

For more information, see Evaluation error codes.