o SiH@sddlZddlmZddlmZmZmZmZmZm Z m Z m Z ddl Z ddlZddlmZddlmZddlmZddlmZddlmZdd lmZmZmZmZmZmZm Z m!Z!eGd d d eZ"dS) N) dataclass)AnyCallableDictIterableListOptionalTupleUnion) Recording) VideoInfo)Cut)FeatureExtractor)SupervisionSegment) LOG_EPSILONPathlikeSecondscompute_num_framescompute_num_samplesfastcopyperturb_num_samplesuuid4c@seZdZUdZeed<eed<eed<eed<dZ e eed<dZ e eed<dZ e eed <dZ e eed <dZe eed <dZe eed <ed efddZeddZed efddZed efddZed efddZed efddZed efddZed efddZded efdd Zed efd!d"Zd efd#d$Zd e e j!fd%d&Z"d e e j!fd'd(Z# )dyd*ed e e$e%j&e e%j&ffd+d,Z'd-dd)d.d/d0ede ed1ed2ed df d3d4Z(d5d.d)d6ded7ed2ed8ed df d9d:Z)ddde*d;d.dfdeded edd?Z. . dzded@edAe ed dfdBdCZ/dydDed@ed dfdEdFZ0dydDed@ed dfdGdHZ1dydDed@ed dfdIdJZ2dd)d.d)dKgddfdLe dMdNedOed@edPe3edQe edRe ed dfdSdTZ4 .d{dUed@ed dfdVdWZ5d|dXdYZ6d|dZd[Z7d|d\d]Z8d|d^d_Z9d|d`daZ:dbe;d e-fdcddZdie?e@ge@fd dfdjdkZAd|dldmZBdne?eCgefd dfdodpZDeEdqed dfdrdsZFdteGd dfdudvZHdteGd dfdwdxZIdS)} PaddingCutaG :class:`~lhotse.cut.PaddingCut` is a dummy :class:`~lhotse.cut.Cut` that doesn't refer to actual recordings or features --it simply returns zero samples in the time domain and a specified features value in the feature domain. Its main role is to be appended to other cuts to make them evenly sized. Please refer to the documentation of :class:`~lhotse.cut.Cut` to learn more about using cuts. See also: - :class:`lhotse.cut.Cut` - :class:`lhotse.cut.MonoCut` - :class:`lhotse.cut.MixedCut` - :class:`lhotse.cut.CutSet` idduration sampling_rate feat_valueN num_frames num_features frame_shift num_samplesvideocustomreturncCdSNrselfr&r&F/home/ubuntu/.local/lib/python3.10/site-packages/lhotse/cut/padding.pystart<zPaddingCut.startcCsgSNr&r'r&r&r) supervisions@r+zPaddingCut.supervisionscCr$r%r&r'r&r&r)channelDr+zPaddingCut.channelcC |jduSr,)rr'r&r&r) has_featuresH zPaddingCut.has_featurescCr/r,r r'r&r&r) has_recordingLr1zPaddingCut.has_recordingcCs|jo|jduSr,)r3r!r'r&r&r) has_videoPszPaddingCut.has_videocCr$Nr&r'r&r&r) num_channelsTr+zPaddingCut.num_channelscCr$)NFr&r'r&r&r) is_in_memoryXr+zPaddingCut.is_in_memoryfieldcCs>|dkr|jS|dkr|jS|dkr|jS|jduo||jvS)N recordingfeaturesr!)r3r0r4r")r(r9r&r&r)has\szPaddingCut.hascCr$)NPADr&r'r&r&r) recording_idfr+zPaddingCut.recording_idcCr$)zEmpty iterable.r&r&r'r&r&r) iter_datajr+zPaddingCut.iter_datacOs&|jrt|j|jftj|jSdSr,)r0nponesrrfloat32rr(argskwargsr&r&r) load_featuresos zPaddingCut.load_featurescOs&|jrtdt|j|jftjSdSr5)r3r@zerosrrrrBrCr&r&r) load_audioxs zPaddingCut.load_audioT with_audiocCsT|jr(d}|rtjdt|j|jtjd}tj|jjd|jj |jj tj d|fSdS)Nr6)dtype) r4torchrGrrrrBr!rheightwidthuint8)r(rIaudior&r&r) load_videos$  zPaddingCut.load_videoF)offsetrkeep_excessive_supervisions preserve_idrSrTrUc Ks|dur |j|n|}|dksJt||r|jntt||j|jdur-t||j|j dnd|j durt ||j ddSddS)a Return a new PaddingCut with region extended by the specified duration. :param duration: The duration by which to extend the cut. :param direction: string, 'left', 'right' or 'both'. Determines whether the cut should be extended to the left, right or both sides. By default, the cut is extended by the specified duration on both sides. :param preserve_id: When ``True``, preserves the cut ID from before padding. Otherwise, generates a new random ID (default). :param pad_silence: See usage in :func:`lhotse.cut.MonoCut.extend_by`. It is ignored here. :return: an extended PaddingCut. r]rRNrVrWrXrY)r(rr^rUr_r[r&r&r) extend_bys.    zPaddingCut.extend_byrightpad_feat_valuepad_value_dictc Cs$ddlm}|||||||||dS)a Return a new MixedCut, padded with zeros in the recording, and ``pad_feat_value`` in each feature bin. The user can choose to pad either to a specific `duration`; a specific number of frames `num_frames`; or a specific number of samples `num_samples`. The three arguments are mutually exclusive. :param duration: The cut's minimal duration after padding. :param num_frames: The cut's total number of frames after padding. :param num_samples: The cut's total number of samples after padding. :param pad_feat_value: A float value that's used for padding the features. By default we assume a log-energy floor of approx. -23 (1e-10 after exp). :param direction: string, 'left', 'right' or 'both'. Determines whether the padding is added before or after the cut. :param preserve_id: When ``True``, preserves the cut ID from before padding. Otherwise, generates a new random ID (default). :param pad_value_dict: Optional dict that specifies what value should be used for padding arrays in custom attributes. :return: a padded MixedCut if duration is greater than this cut's duration, otherwise ``self``. r6)pad)rrr rbr^rUrc)setrd) r(rrr rbr^rUrcrdr&r&r)rds zPaddingCut.padaffix_idrecording_fieldc CsB|jsJdt||r|jd|n|j|t|j|ddddS)a  Return a new ``PaddingCut`` that will "mimic" the effect of resampling on ``sampling_rate``, ``duration``, and ``num_samples``. :param sampling_rate: The new sampling rate. :param affix_id: Should we modify the ID (useful if both versions of the same cut are going to be present in a single manifest). :param recording_field: which recording field to resample. Ignored, present for interface compatibility. :return: a modified copy of the current ``PaddingCut``. z/Cannot resample a PaddingCut without Recording._rsN)rrr rrr)r3rrrr)r(rrfrgr&r&r)resample s zPaddingCut.resamplefactorc Cr|jrtdd}d}d}n |j}|j}|j}t|j|}||j}t ||r/|j d|n|j |||||dS)a Return a new ``PaddingCut`` that will "mimic" the effect of speed perturbation on ``duration`` and ``num_samples``. :param factor: The speed will be adjusted this many times (e.g. factor=1.1 means 1.1x faster). :param affix_id: When true, we will modify the ``PaddingCut.id`` field by affixing it with "_sp{factor}". :return: a modified copy of the current ``PaddingCut``. zAttempting to perturb speed on a MonoCut that references pre-computed features. The feature manifest will be detached, as we do not support feature-domain speed perturbation.N_sprr rrrr r0loggingwarningrrrrr rrrr(rjrfnew_num_framesnew_num_featuresnew_frame_shiftnew_num_samplesr[r&r&r) perturb_speed&s*   zPaddingCut.perturb_speedc Crk)a Return a new ``PaddingCut`` that will "mimic" the effect of tempo perturbation on ``duration`` and ``num_samples``. Compared to speed perturbation, tempo preserves pitch. :param factor: The tempo will be adjusted this many times (e.g. factor=1.1 means 1.1x faster). :param affix_id: When true, we will modify the ``PaddingCut.id`` field by affixing it with "_tp{factor}". :return: a modified copy of the current ``PaddingCut``. zAttempting to perturb tempo on a MonoCut that references pre-computed features. The feature manifest will be detached, as we do not support feature-domain tempo perturbation.N_tprmrnrqr&r&r) perturb_tempoJs*   zPaddingCut.perturb_tempocC&t||r|jd|dS|jdS)a Return a new ``PaddingCut`` that will "mimic" the effect of volume perturbation on amplitude of samples. :param factor: The volume will be adjusted this many times (e.g. factor=1.1 means 1.1x louder). :param affix_id: When true, we will modify the ``PaddingCut.id`` field by affixing it with "_vp{factor}". :return: a modified copy of the current ``PaddingCut``. _vprrr)r(rjrfr&r&r)perturb_volumeos& zPaddingCut.perturb_volumer rir_recordingr normalize_output early_only rir_channels room_rng_seedsource_rng_seedcCs"t||r |jddS|jdS)a Return a new ``PaddingCut`` that will "mimic" the effect of reverberation with impulse response on original samples. :param rir_recording: The impulse response to use for convolving. :param normalize_output: When true, output will be normalized to have energy as input. :param early_only: When true, only the early reflections (first 50 ms) will be used. :param affix_id: When true, we will modify the ``PaddingCut.id`` field by affixing it with "_rvb". :param rir_channels: The channels of the impulse response to use. :return: a modified copy of the current ``PaddingCut``. _rvbr{r|)r(r~rrrfrrrr&r&r) reverb_rir|s"zPaddingCut.reverb_rirtargetcKry)aQ Return a new ``PaddingCut`` that will "mimic" the effect of loudness normalization :param target: The target loudness in dBFS. :param affix_id: When true, we will modify the ``DataCut.id`` field by affixing it with "_ln{target}". :return: a modified copy of the current ``DataCut``. _lnr{r|)r(rrfrEr&r&r)normalize_loudnesss  zPaddingCut.normalize_loudnesscCs(|js Jd|jdt|ddddS)zNReturn a copy of the current :class:`.PaddingCut`, detached from ``features``.zBCannot detach features from a MonoCut with no Recording (cut ID = ).N)rrr)r3rrr'r&r&r) drop_featuress  zPaddingCut.drop_featurescCs$|js Jd|jdt|ddS)zOReturn a copy of the current :class:`.PaddingCut`, detached from ``recording``.zECannot detach recording from a PaddingCut with no Features (cut ID = rNr2)r0rrr'r&r&r)drop_recordings   zPaddingCut.drop_recordingcC|SzNo-opr&r'r&r&r)drop_supervisionsr+zPaddingCut.drop_supervisionscCrrr&r'r&r&r)drop_alignmentsr+zPaddingCut.drop_alignmentscCr)zNo-op.r&r'r&r&r)drop_in_memory_datar+zPaddingCut.drop_in_memory_data extractorcOs*t|||jt|j|j|jd|jdS)z Returns a new PaddingCut with updates information about the feature dimension and number of feature frames, depending on the ``extractor`` properties. rV)rrr)r feature_dimrrrr)r(rrDrEr&r&r)compute_and_store_featuress z%PaddingCut.compute_and_store_featurescOrzT Just for consistency with :class`.MonoCut` and :class:`.MixedCut`. r&rCr&r&r)fill_supervisionzPaddingCut.fill_supervisioncOrrr&rCr&r&r)move_to_memoryrzPaddingCut.move_to_memory transform_fncCr)z Just for consistency with :class:`.MonoCut` and :class:`.MixedCut`. :param transform_fn: a dummy function that would be never called actually. :return: the PaddingCut itself. r&)r(rr&r&r)map_supervisionsszPaddingCut.map_supervisionscOr)z~ Just for consistency with :class:`.MonoCut` and :class:`.MixedCut`. :return: the PaddingCut itself. r&rCr&r&r)merge_supervisionsszPaddingCut.merge_supervisions predicatecCr)z Just for consistency with :class:`.MonoCut` and :class:`.MixedCut`. :param predicate: A callable that accepts `SupervisionSegment` and returns bool :return: a modified MonoCut r&)r(rr&r&r)filter_supervisionss zPaddingCut.filter_supervisionsdatacCs|ddtdi|S)Ntyper&)popr)rr&r&r) from_dicts zPaddingCut.from_dictpathcCrr,r&r(rr&r&r)with_features_path_prefixz$PaddingCut.with_features_path_prefixcCrr,r&rr&r&r)with_recording_path_prefixrz%PaddingCut.with_recording_path_prefix)T)FN)F)r#r)J__name__ __module__ __qualname____doc__rZ__annotations__rintfloatrrrrr r!r r"dictpropertyr*r-r.boolr0r3r4r7r8r<r>rr?r@ndarrayrFrHr rLTensorrQr\r`rrr r rdrirvrxr}rrrrrrrrrrrrrrrrrr staticmethodrrrrr&r&r&r)rsD       # - - $%            r)#ro dataclassesrtypingrrrrrrr r numpyr@rL lhotse.audior lhotse.audio.utilsr lhotse.cut.baser lhotse.featuresrlhotse.supervisionr lhotse.utilsrrrrrrrrrr&r&r&r)s (     (