o i@ @svddlZddlZddlZddlZddlZddlZddlZddlZ ddl m Z ddl m Z ddl mZm Z ddlmZmZddlmmZddlmZdefdd Zd e ejeffd d Zd-dedefddZddZdededejdejdef ddZddZddZ d.ddZ!d d!Z"Gd"d#d#ed$Z#Gd%d&d&e#Z$Gd'd(d(e#Z%Gd)d*d*e#Z&iZ'Gd+d,d,e(Z)dS)/N)Union)Path) Generatorr)ABCMetaabstractmethod)pad_listaudiocCsd}tt|D]W}||}|d|dksJdt|dtjs&Jdt|ddt|dd|djdksAJd|dkrU|d||ddksUJd||d|d7}q|S)Nrz$modelscope error: Wrong time stamps.z"modelscope error: Wrong data type.>zFmodelscope error: audio data in list is inconsistent with time length.)rangelen isinstancenpndarrayintshape)r audio_durisegrP/home/ubuntu/.local/lib/python3.10/site-packages/funasr/models/campplus/utils.pycheck_audio_lists0 rinputscCsg}tt|D]s}t||trAt||}tjt |dd\}}t|j dkr3|dddf}t | d}|d}n5t||tjrrt||j dksVJd||}|jdvrg|dd}n|d}t |}ntd ||q|S) Nfloat32)dtyper rr z.modelscope error: Input array should be [N, T])int16int32int64izOmodelscope error: The input type is restricted to audio address and nump array.)r r rstrFilereadsfloadioBytesIOrtorch from_numpy unsqueezesqueezerrrastype ValueErrorappend)routputr file_bytesdatafsrrr sv_preprocess(s*     r1r vad_segmentsreturncs@dddfdd}g}t|D] \}}|||q|S)Ng?g?)seg_dur seg_shiftc s|d}|d}td}td}d}g}td|jd|D]G}t|||jd}||kr8|S|}td||}|||} | jd|kr\t| d|| jdfd} |||||| gq%|S)Nrr r4r5constant)rr rminmaxrpadr,) seg_dataseg_str/ chunk_len chunk_shift last_chunk_edseg_reschunk_stchunk_ed chunk_dataconfigr0rr seg_chunkHs$ "zsv_chunk..seg_chunk) enumerateextend)r2r0rEsegsrsrrCrsv_chunkBsrJcCs|g}g}g}|D]*}tj|ddd}||jddd}||||jd||jdqt|dd}|||fS)NrP) num_mel_binsT)dimkeepdim) pad_value)Kaldifbankr(meanr,rr)rfeatures feature_timesfeature_lengthsaufeaturefeatures_paddedrrrextract_featurebs   rYsegmentslabels embeddingsc Cs t|t|ks Jt|}g}tt|D]}|||d||d||gqt|}g}t|dD]}|||kd}||q9t|}dd}tdt|D]/}|||dd||dr||d||ddd} | ||d<| ||dd<qZt |}|S)Nrr cSs||dkrdSdS)Ng-C6?TFr)t1t2rrr is_overlappeds z"postprocess..is_overlappedr ) r correct_labelsr r, merge_sequer8rRrstacksmooth) rZr2r[r\distribute_resrspk_embsspk_embr_prrr postprocessrs(&    rhcCsFd}i}g}|D]}||vr|||<|d7}|||qt|S)Nrr )r,rarray)r[ labels_idid2id new_labelsrrrrr`s r`cCsv|dg}tdt|D],}||d|ddks&||d|ddkr.|||q ||d|dd<q |S)Nrr r )r r r,)rdresrrrrras 0raffffff?cCs4t|dkr|Stt|D]}t||dd||d<t||dd||d<||d||d|kr|dkrI||dd||d<q|t|dkr^||dd||d<q||d||dd||dd||dkr||dd||d<q||dd||d<qt|}|S)Nr rr )r r roundra)rnmindurrrrrrcs 8rcc Csdd|D}|D]=}|d}|d}d}d}|D]&\}}} tt||t||d} | |kr3| }| }| dkr?|| kr?|| 7}qt||d<q |S)NcSs$g|]\}}}|d|d|fqS)ir).0spk_stspk_edspkrrr s$z"distribute_spk..startendrru)r8r7r) sentence_list sd_time_listdsentence_start sentence_end sentence_spk max_overlaprsrtruoverlaprrrdistribute_spks rc @s~eZdZdZedefddZedefddZedede ee fdd fd d Z eddede ee fd edd fddZ d S)StoragezAbstract class of storage. All backends need to implement two apis: ``read()`` and ``read_text()``. ``read()`` reads the file as a byte stream and ``read_text()`` reads the file as texts. filepathcCdSNrselfrrrrr!z Storage.readcCrrrrrrr read_textrzStorage.read_textobjr3NcCrrrrrrrrrwriterz Storage.writeutf-8encodingcCrrrrrrrrrr write_textrzStorage.write_textr) __name__ __module__ __qualname____doc__rrr!rbytesrrrrrrrrrs *r) metaclassc @seZdZdZdeeefdefddZddeeefdedefdd Z d edeeefdd fd d Z dd edeeefdedd fddZ e j deeefdeeeefd d ffddZd S) LocalStoragezLocal hard disk storagerr3cCs8t|d }|}Wd|S1swY|S)Read data from a given ``filepath`` with 'rb' mode. Args: filepath (str or Path): Path to read data. Returns: bytes: Expected bytes object. rbNopenr!)rrfcontentrrrr!s  zLocalStorage.readrrcCs<t|d|d }|}Wd|S1swY|S)>Read data from a given ``filepath`` with 'r' mode. Args: filepath (str or Path): Path to read data. encoding (str): The encoding format used to open the ``filepath``. Default: 'utf-8'. Returns: str: Expected text reading from ``filepath``. rrNr)rrrr value_bufrrrrs   zLocalStorage.read_textrNcCsdtj|}|rtj|stj|ddt|d}||WddS1s+wYdS))Write data to a given ``filepath`` with 'wb' mode. Note: ``write`` will create a directory if the directory of ``filepath`` does not exist. Args: obj (bytes): Data to be written. filepath (str or Path): Path to write data. Texist_okwbNospathdirnameexistsmakedirsrr)rrrrrrrrr s   "zLocalStorage.writecCshtj|}|rtj|stj|ddt|d|d}||WddS1s-wYdS)Write data to a given ``filepath`` with 'w' mode. Note: ``write_text`` will create a directory if the directory of ``filepath`` does not exist. Args: obj (str): Data to be written. filepath (str or Path): Path to write data. encoding (str): The encoding format used to open the ``filepath``. Default: 'utf-8'. TrwrNr)rrrrrrrrrrs  "zLocalStorage.write_textccs |VdSz$Only for unified API and do nothing.Nrrrrr as_local_path3s zLocalStorage.as_local_pathr)rrrrrrrrr!rrr contextlibcontextmanagerrrrrrrrs $2rc @seZdZdZddZddZejdede e ee fddffd d Z d e d e ee fddfd dZdd ed e ee fdeddfddZdS) HTTPStoragezHTTP and HTTPS storage.cCt|}||jSr)requestsgetraise_for_statusrrurlrrrrr!<s zHTTPStorage.readcCrr)rrrtextrrrrrBs zHTTPStorage.read_textrr3Nc cPztjdd}|||||jVWt|jdSt|jw)ahDownload a file from ``filepath``. ``as_local_path`` is decorated by :meth:`contextlib.contextmanager`. It can be called with ``with`` statement, and when exists from the ``with`` statement, the temporary path will be released. Args: filepath (str): Download a file from ``filepath``. Examples: >>> storage = HTTPStorage() >>> # After existing from the ``with`` clause, >>> # the path will be removed >>> with storage.get_local_path('http://path/to/file') as path: ... # do something here FdeleteNtempfileNamedTemporaryFilerr!closenamerremoverrrrrrrG  zHTTPStorage.as_local_pathrrcCtd)Nz&write is not supported by HTTP StorageNotImplementedError)rrrrrrrazHTTPStorage.writerrcCr)Nz+write_text is not supported by HTTP Storager)rrrrrrrrdrzHTTPStorage.write_textr)rrrrr!rrrrrrrrrrrrrrrr9s&(rc @seZdZdZdddZddZddd Zejd e d e e e e fddffd d Z ded e e e fd dfddZdde d e e e fde d dfddZdS) OSSStoragez OSS storage.NcCr)Nz3OSSStorage.__init__ to be implemented in the futurer)ross_config_filerrr__init__kszOSSStorage.__init__cCr)Nz/OSSStorage.read to be implemented in the futurerrrrrr!orzOSSStorage.readrcCr)Nz4OSSStorage.read_text to be implemented in the futurer)rrrrrrrrrzOSSStorage.read_textrr3c cr)agDownload a file from ``filepath``. ``as_local_path`` is decorated by :meth:`contextlib.contextmanager`. It can be called with ``with`` statement, and when exists from the ``with`` statement, the temporary path will be released. Args: filepath (str): Download a file from ``filepath``. Examples: >>> storage = OSSStorage() >>> # After existing from the ``with`` clause, >>> # the path will be removed >>> with storage.get_local_path('http://path/to/file') as path: ... # do something here FrNrrrrrrurzOSSStorage.as_local_pathrcCr)Nz0OSSStorage.write to be implemented in the futurerrrrrrrzOSSStorage.writercCr)Nz5OSSStorage.write_text to be implemented in the futurerrrrrrrzOSSStorage.write_textrr)rrrrrr!rrrrrrrrrrrrrrrrhs  &(rc @seZdZUeeeedZeed<e ddZ e de de fddZ e ddee efd e de fd d Ze d e dee efddfddZe dd e de d e ddfddZejde deee efddffddZdS)r )osshttphttpslocal_prefix_to_storagecCst|tsJdt|d|vrd}n |d\}}|}|tjvs2Jd|dttj|tvr>tj|t|<t|S)Nz uri should be str type, but got z://rzUnsupported uri z, valid prefixs: ) rrtypesplitr rlistkeys G_STORAGES)uri storage_typeprefix_rrr _get_storages zFile._get_storagerr3cCst|}||S)r)r rr!)rstoragerrrr!s z File.readrrcCst|}||S)r)r rr)rrrrrrrs zFile.read_textrNcCst|}|||S)r)r rr)rrrrrrrs z File.writecCst|}|||S)r)r rr)rrrrrrrrs  zFile.write_textccsBt|}|| }|VWddS1swYdSr)r rr)rr local_pathrrrrs   "zFile.as_local_pathr)rrrrrrrdict__annotations__ staticmethodrrrr!rrrrrrrrrrrrrr s$   " *r )r )ro)*r$rr&rrrnumpyrlibrosar"typingrpathlibrrabcrrtorchaudio.compliance.kaldi compliancekaldirP*funasr.models.transformer.utils.nets_utilsrrrrr1rJrYrhr`rarcrrrrrrobjectr rrrrsN     $ K/.