qi"dZddlmZddlmZmZmZmZddlZddl m Z ddl m Z ddl mZmZmZmZddlmZmZdd lmZdd lmZmZdd lmZd ed eefdZeGddeZGddeZy)zXTTS text-to-speech service implementation. This module provides integration with Coqui XTTS streaming server for text-to-speech synthesis using local Docker deployment. ) dataclass)AnyAsyncGeneratorDictOptionalN)logger)create_stream_resampler) ErrorFrameFrame StartFrameTTSAudioRawFrame) TTSSettings_warn_deprecated_param) TTSService)Languageresolve_language) traced_ttslanguagereturncBitjdtjdtjdtjdtj dtj dtjdtjdtjd tjd tjd tjd tjd tjdtjdtj d}t#||dS)zConvert a Language enum to XTTS language code. Args: language: The Language enum value to convert. Returns: The corresponding XTTS language code, or None if not supported. csdeenesfrhihuitjakonlplptrutrzzh-cnT) use_base_code)rCSDEENESFRHIHUITJAKONLPLPTRUTRZHr)r LANGUAGE_MAPs K/opt/pipecat/venv/lib/python3.12/site-packages/pipecat/services/xtts/tts.pylanguage_to_xtts_languager9's T T  T  T   T   T   T  T  T  T  T  T  T  T  T"  W#L( Hl$ GGceZdZdZy)XTTSTTSSettingszSettings for XTTSService.N)__name__ __module__ __qualname____doc__r:r8r<r<Gs#r:r<ceZdZUdZeZeed<dejdddde e de de jded e e d e ef fd Zd efd Zded e e fdZdeffd Zede de d eedffdZxZS) XTTSServicezCoqui XTTS text-to-speech service. Provides text-to-speech synthesis using a locally running Coqui XTTS streaming server. Supports multiple languages and voice cloning through studio speakers configuration. _settingsN)voice_idr sample_ratesettingsrEbase_urlaiohttp_sessionrrFrGc tdd|j|}|tdtd||_||j |t |d|dd|d|||_d|_||_ t|_ y)aInitialize the XTTS service. Args: voice_id: ID of the voice/speaker to use for synthesis. .. deprecated:: 0.0.105 Use ``settings=XTTSTTSSettings(voice=...)`` instead. base_url: Base URL of the XTTS streaming server. aiohttp_session: HTTP session for making requests to the server. language: Language for synthesis. Defaults to English. sample_rate: Audio sample rate. If None, uses default. settings: Runtime-updatable settings. When provided alongside deprecated parameters, ``settings`` values take precedence. **kwargs: Additional arguments passed to parent TTSService. N)modelvoicerrErLT)rFpush_start_framepush_stop_framesrGrA) r<language_to_service_languagerrL apply_updatesuper__init__ _base_url_studio_speakers_aiohttp_sessionr _resampler) selfrErHrIrrFrGkwargsdefault_settings __class__s r8rRzXTTSService.__init__Ys8+66x@    ": H%-  "    ) )( 3  #!!%    ":> /13r:rcy)zCheck if this service can generate processing metrics. Returns: True, as XTTS service supports metrics generation. TrA)rWs r8can_generate_metricsz XTTSService.can_generate_metricss r:ct|S)zConvert a Language enum to XTTS service language format. Args: language: The language to convert. Returns: The XTTS-specific language code, or None if not supported. )r9)rWrs r8rOz(XTTSService.language_to_service_languages)22r:framec Kt||d{|jry|jj |j dz4d{}|j dk7rU|jd{}|jd|j d|dd{ dddd{y|jd{|_dddd{y777s7J7;7$7#1d{7swYyxYww)zStart the XTTS service and load studio speakers. Args: frame: The start frame containing initialization parameters. Nz/studio_speakersz'Error getting studio speakers (status: , error: )) error_msg) rQstartrTrUgetrSstatustext push_errorjson)rWr^rrgrZs r8rdzXTTSService.starts  gmE"""  ((,,T^^>P-PQ 3 3UVxx3VVX~oo GzQZ[_Z``ab&  3 3 3+,&&(ND ! 3 3 3 #  3% 3%3 3 3 3 3sDC+=DC-D#C9;C/<*C9&C1'C9, D7C38D=C9C5 C9 D%C7&D-D/C91C93D5C97D9D ?D D Drg context_idcKtj|d|d|jstj|dy|j|jj }|j dz}|jddjdd|jj|d |d d d d }|jj||4d{}|jdk7rJ|jd{}td|jd|d dddd{y|j|d{|j}t!}|j"j%|23d{} t'| dkDs|j)d{|j+| t'|dk\sP|dd} |dd}|j,j/t1| d|j2d{} t5| |j2d|} | t'|dk\ro7h7D77777A6t'|dkDrW|j,j/t1|d|j2d{7} t5| |j2d|} | dddd{7y#1d{7swYyxYww)aGenerate speech from text using XTTS streaming server. Args: text: The text to synthesize into speech. context_id: The context ID for tracking audio frames. Yields: Frame: Audio frames containing the synthesized speech. z: Generating TTS []z no studio speakers availableNz /tts_stream.*speaker_embeddinggpt_cond_latentF)rgrrqrradd_wav_headerstream_chunk_size)rir`zError getting audio (status: rarb)errorrii])rk)rdebugrTrvrDrLrSreplacerrUpostrfrgr start_tts_usage_metrics chunk_size bytearraycontent iter_chunkedlenstop_ttfb_metricsextendrVresamplebytesrFr ) rWrgrk embeddingsurlpayloadrj CHUNK_SIZEbufferchunk process_dataresampled_audior^s r8run_ttszXTTSService.run_ttss  v/vQ78$$ LLD6!>? @ **4>>+?+?@ nn},LLb)11#r://!+,?!@)*;<#!#  ((--c-@, , Axx3VVX~ )FqxxjPYZ^Y__`'abb , , ,  ..t4 4 4J[F yy55jA $ $eu:>00222MM%(F u,(.fu~ !'150H0H!,/8H8H1+!1+T-=-=qZ!$ F u,', %,  5  $2+! B46{Q(,(@(@&M5$*:*:)##)#T%5%5qZ Y, , , , , sCKIK#K I%K & K1I2K7K  I  5K II I K K ,I-#K =K I0K K KK K K  IK K AK J#K 8 KKK KK KK)r=r>r?r@r<Settings__annotations__rr)rstraiohttp ClientSessionintrRboolr\rOr rdrrr r __classcell__)rZs@r8rCrCNsH #'&[[%).2;43-;4 ;4 !.. ;4  ;4c];4?+;4zd 3X 3(3- 333(I#I3I>%QU+;VIIr:rC) r@ dataclassesrtypingrrrrrlogururpipecat.audio.utilsr pipecat.frames.framesr r r r pipecat.services.settingsrrpipecat.services.tts_servicerpipecat.transcriptions.languagerr(pipecat.utils.tracing.service_decoratorsrrr9r<rCrAr:r8rs{ "667 J3F?HHXc]H@  k   y*yr: