TFDS תומך כעת בפורמט קרואסון 🥐 ! קרא את התיעוד כדי לדעת יותר.

דף זה תורגם על ידי Cloud Translation API.

openslr

הפניות:

SLR41

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR41')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	5822

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR42

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR42')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	2906

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR43

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR43')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	2064

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR44

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR44')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	4213

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR63

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR63')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	4126

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR64

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR64')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	1569

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR65

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR65')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	4284

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR66

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR66')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	4448

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR69

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR69')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	4240

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR35

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR35')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	185076

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR36

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR36')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	219156

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR70

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR70')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	3359

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR71

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR71')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	4374

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR72

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR72')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	4903

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR73

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR73')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	5447

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR74

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR74')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	617

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR75

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR75')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	3357

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR76

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR76')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	7136

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR77

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR77')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	5587

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR78

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR78')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	4272

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR79

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR79')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	4400

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR80

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR80')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	2530

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR86

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR86')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	3583

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR32

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR32')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	9821

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "struct",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR52

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR52')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	185293

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR53

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR53')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	218703

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR54

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR54')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	157905

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

SLR83

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:openslr/SLR83')

תיאור :

OpenSLR is a site devoted to hosting speech and language resources, such as training corpora for speech recognition,
and software related to speech recognition. We intend to be a convenient place for anyone to put resources that
they have created, so that they can be downloaded publicly.

רישיון : אין רישיון ידוע
גרסה : 0.0.0
פיצולים :

לְפַצֵל	דוגמאות
`'train'`	17877

תכונות :

{
    "path": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "audio": {
        "sampling_rate": 48000,
        "mono": true,
        "_storage_dtype": "string",
        "id": null,
        "_type": "Audio"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}