Module audiocraft.data.info_audio_dataset

Base classes for the datasets that also provide non-audio metadata, e.g. description, text transcription etc.

Functions

def clusterify_all_meta(meta: List[AudioMeta]) ‑> List[AudioMeta]
Expand source code 
def clusterify_all_meta(meta: tp.List[AudioMeta]) -> tp.List[AudioMeta]:
    """Monkey-patch all meta to match cluster specificities."""
    return [_clusterify_meta(m) for m in meta]

Monkey-patch all meta to match cluster specificities.

def get_keyword(value: str | None) ‑> str | None
Expand source code 
def get_keyword(value: tp.Optional[str]) -> tp.Optional[str]:
    """Preprocess a single keyword."""
    if value is None or (not isinstance(value, str)) or len(value) == 0 or value == 'None':
        return None
    else:
        return value.strip().lower()

Preprocess a single keyword.

def get_keyword_list(values: str | List[str]) ‑> List[str] | None
Expand source code 
def get_keyword_list(values: tp.Union[str, tp.List[str]]) -> tp.Optional[tp.List[str]]:
    """Preprocess a list of keywords."""
    if isinstance(values, str):
        values = [v.strip() for v in re.split(r'[,\s]', values)]
    elif isinstance(values, float) and math.isnan(values):
        values = []
    if not isinstance(values, list):
        logger.debug(f"Unexpected keyword list {values}")
        values = [str(values)]

    kws = [get_keyword(v) for v in values]
    kw_list = [k for k in kws if k is not None]
    if len(kw_list) == 0:
        return None
    else:
        return kw_list

Preprocess a list of keywords.

def get_keyword_or_keyword_list(value: str | None) ‑> str | List[str] | None
Expand source code 
def get_keyword_or_keyword_list(value: tp.Optional[str]) -> tp.Union[tp.Optional[str], tp.Optional[tp.List[str]]]:
    """Preprocess a single keyword or possible a list of keywords."""
    if isinstance(value, list):
        return get_keyword_list(value)
    else:
        return get_keyword(value)

Preprocess a single keyword or possible a list of keywords.

def get_string(value: str | None) ‑> str | None
Expand source code 
def get_string(value: tp.Optional[str]) -> tp.Optional[str]:
    """Preprocess a single keyword."""
    if value is None or (not isinstance(value, str)) or len(value) == 0 or value == 'None':
        return None
    else:
        return value.strip()

Preprocess a single keyword.

Classes

class AudioInfo (meta: AudioMeta,
seek_time: float,
n_frames: int,
total_frames: int,
sample_rate: int,
channels: int,
audio_tokens: torch.Tensor | None = None)
Expand source code 
@dataclass
class AudioInfo(SegmentWithAttributes):
    """Dummy SegmentInfo with empty attributes.

    The InfoAudioDataset is expected to return metadata that inherits
    from SegmentWithAttributes class and can return conditioning attributes.

    This basically guarantees all datasets will be compatible with current
    solver that contain conditioners requiring this.
    """
    audio_tokens: tp.Optional[torch.Tensor] = None  # populated when using cached batch for training a LM.

    def to_condition_attributes(self) -> ConditioningAttributes:
        return ConditioningAttributes()

Dummy SegmentInfo with empty attributes.

The InfoAudioDataset is expected to return metadata that inherits from SegmentWithAttributes class and can return conditioning attributes.

This basically guarantees all datasets will be compatible with current solver that contain conditioners requiring this.

Ancestors

Subclasses

Class variables

var audio_tokens : torch.Tensor | None

Methods

def to_condition_attributes(self) ‑> ConditioningAttributes
Expand source code 
def to_condition_attributes(self) -> ConditioningAttributes:
    return ConditioningAttributes()
class InfoAudioDataset (meta: List[AudioMeta],
**kwargs)
Expand source code 
class InfoAudioDataset(AudioDataset):
    """AudioDataset that always returns metadata as SegmentWithAttributes along with the audio waveform.

    See `audiocraft.data.audio_dataset.AudioDataset` for initialization arguments.
    """
    def __init__(self, meta: tp.List[AudioMeta], **kwargs):
        super().__init__(clusterify_all_meta(meta), **kwargs)

    def __getitem__(self, index: int) -> tp.Union[torch.Tensor, tp.Tuple[torch.Tensor, SegmentWithAttributes]]:
        if not self.return_info:
            wav = super().__getitem__(index)
            assert isinstance(wav, torch.Tensor)
            return wav
        wav, meta = super().__getitem__(index)
        return wav, AudioInfo(**meta.to_dict())

AudioDataset that always returns metadata as SegmentWithAttributes along with the audio waveform.

See AudioDataset for initialization arguments.

Ancestors

Subclasses

Inherited members