o i#@sdZddlmZddlmZmZmZmZddlZddl m Z ddl m Z ddl mZmZmZmZddlmZdd lmZdd lmZmZdd lmZd ed eefddZeGdddeZGdddeZdS)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) TTSService)Languageresolve_language) traced_ttslanguagereturncCsitjdtjdtjdtjdtjdtjdtjdtjdtj d tj d tj d tj d tj d tjdtjdtjd}t||ddS)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_MAPr5M/home/ubuntu/.local/lib/python3.10/site-packages/pipecat/services/xtts/tts.pylanguage_to_xtts_language'sD      r7c@seZdZdZdS)XTTSTTSSettingszSettings for XTTSService.N)__name__ __module__ __qualname____doc__r5r5r5r6r8Gsr8cseZdZUdZeZeed<dejdddde e de de j ded e e d e ef fd d Zd efddZded e e fddZdeffdd Zede de d eedffddZZS) 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_ratesettingsr?base_urlaiohttp_sessionrr@rAc  s|jddtjd}|dur|dd||_|dur#|dd||_|dur,||tjd|dd|d|||_ d|_ ||_ t |_ dS) aYInitialize the XTTS service. Args: voice_id: ID of the voice/speaker to use for synthesis. .. deprecated:: 0.0.105 Use ``settings=XTTSService.Settings(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. .. deprecated:: 0.0.106 Use ``settings=XTTSService.Settings(language=...)`` instead. 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)modelvoicerr?rErT)r@push_start_framepush_stop_framesrAr5)Settingsrr&"_warn_init_param_moved_to_settingsrEr apply_updatesuper__init__ _base_url_studio_speakers_aiohttp_sessionr _resampler) selfr?rBrCrr@rAkwargsdefault_settings __class__r5r6rLYs2      zXTTSService.__init__rcCsdS)zCheck if this service can generate processing metrics. Returns: True, as XTTS service supports metrics generation. Tr5)rQr5r5r6can_generate_metricssz XTTSService.can_generate_metricscCst|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. )r7)rQrr5r5r6language_to_service_languages z(XTTSService.language_to_service_languageframec st|IdH|jrdS|j|jd4IdH=}|jdkrF|IdH}|jd|jd|ddIdH WdIdHdS| IdH|_WdIdHdS1IdHs_wYdS)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) rKstartrNrOgetrMstatustext push_errorjson)rQrXrr`rTr5r6r]s  .zXTTSService.startr` context_idc Cs&t|d|d|jst|ddS|j|jj}|jd}|dddd|jj|d |d d d d }|j j ||d4IdH}|j dkro| IdH}t d|j d|ddV WdIdHdS||IdH|j}t}|j|2zJ3dHW} t| dkr|IdH|| t|dkr|dd} |dd}|jt| d|jIdH} t| |jd|d} | Vt|dksq6t|dkr|jt|d|jIdH} t| |jd|d} | VWdIdHdSWdIdHdS1IdHs wYdS)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)r`rrirjadd_wav_headerstream_chunk_size)rbrYzError getting audio (status: rZr[)errorrii])rd)rdebugrNrnr>rErMreplacerrOpostr_r`r start_tts_usage_metrics chunk_size bytearraycontent iter_chunkedlenstop_ttfb_metricsextendrPresamplebytesr@r ) rQr`rd embeddingsurlpayloadrc CHUNK_SIZEbufferchunk process_dataresampled_audiorXr5r5r6run_ttssf                %0zXTTSService.run_tts)r9r:r;r<r8rH__annotations__rr&rstraiohttp ClientSessionintrLboolrVrWr r]rrr r __classcell__r5r5rTr6r=Ns4 D (r=)r< dataclassesrtypingrrrrrlogururpipecat.audio.utilsrpipecat.frames.framesr r r r pipecat.services.settingsr pipecat.services.tts_servicerpipecat.transcriptions.languagerr(pipecat.utils.tracing.service_decoratorsrrr7r8r=r5r5r5r6s