covost2

Người giới thiệu:

en_de

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_de')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_tr

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_tr')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_fa

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_fa')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_sv-SE

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_sv-SE')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_mn

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_mn')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_zh-CN

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_zh-CN')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_cy

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_cy')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_ca

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_ca')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_sl

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_sl')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_et

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_et')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_id

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_id')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_ar

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_ar')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_ta

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_ta')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_lv

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_lv')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

en_ja

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/en_ja')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 15531
'train' 289430
'validation' 15531
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

bạn_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/fr_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 14760
'train' 207374
'validation' 14760
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

de_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/de_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 13511
'train' 127834
'validation' 13511
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

es_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/es_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 13221
'train' 79015
'validation' 13221
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ca_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/ca_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 12730
'train' 95854
'validation' 12730
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

nó_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/it_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 8951
'train' 31698
'validation' 8940
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ru_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/ru_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 6300
'train' 12112
'validation' 6110
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

zh-CN_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/zh-CN_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 4898
'train' 7085
'validation' 4843
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

pt_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/pt_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 4023
'train' 9158
'validation' 3318
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

fa_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/fa_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 3445
'train' 53949
'validation' 3445
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

et_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/et_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 1571
'train' 1782
'validation' 1576
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

mn_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/mn_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 1759
'train' 2067
'validation' 1761
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

nl_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/nl_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 1699
'train' 7108
'validation' 1699
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tr_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/tr_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 1629
'train' 3966
'validation' 1624
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ar_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/ar_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 1695
'train' 2283
'validation' 1758
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

sv-SE_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/sv-SE_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 1595
'train' 2160
'validation' 1349
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

lv_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/lv_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 1629
'train' 2337
'validation' 1125
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

sl_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/sl_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 360
'train' 1843
'validation' 509
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ta_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/ta_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 786
'train' 1358
'validation' 384
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ja_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/ja_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 684
'train' 1119
'validation' 635
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

id_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/id_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 844
'train' 1243
'validation' 792
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

cy_en

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:covost2/cy_en')
  • Sự miêu tả :
CoVoST 2, a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. The dataset is created using Mozilla’s open source Common Voice database of crowdsourced voice recordings.

Note that in order to limit the required storage for preparing this dataset, the audio
is stored in the .mp3 format and is not converted to a float32 array. To convert, the audio
file to a float32 array, please make use of the `.map()` function as follows:


python
import torchaudio

def map_to_array(batch):
    speech_array, _ = torchaudio.load(batch["file"])
    batch["speech"] = speech_array.numpy()
    return batch

dataset = dataset.map(map_to_array, remove_columns=["file"])
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 1.0.0
  • Chia tách :
Tách ra Ví dụ
'test' 690
'train' 1241
'validation' 690
  • Đặc trưng :
{
    "client_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "file": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}