o iF @s"dZddlZddlmZddlmZmZmZddlZddl m Z ddl m Z m Z mZmZmZmZmZddlmZddlmZmZdd lmZzdd lmZdd lmZWn eytZ ze !d e e !d e"de dZ [ wweGdddeZ#GdddeZ$GdddeZ%dS)zDeepgram text-to-speech service implementation. This module provides integration with Deepgram's text-to-speech API for generating speech from text using various voice models. N) dataclass)AnyAsyncGeneratorOptional)logger) CancelFrameEndFrame ErrorFrameFrame StartFrameTTSAudioRawFrameTTSStoppedFrame) TTSSettings) TTSServiceWebsocketTTSService) traced_tts)connect)Statez Exception: z\In order to use DeepgramWebsocketTTSService, you need to `pip install pipecat-ai[deepgram]`.zMissing module: c@seZdZdZdS)DeepgramTTSSettingsz;Settings for DeepgramTTSService and DeepgramHttpTTSService.N)__name__ __module__ __qualname____doc__rrQ/home/ubuntu/.local/lib/python3.10/site-packages/pipecat/services/deepgram/tts.pyr,srcsDeZdZUdZeZeed<dZddddddded e ed ed e e d ed e ef fddZ de fddZ deffdd Zdeffdd Zdeffdd ZfddZfddZdedeeefffdd Zd!d"Zd#d$Zd%d&Zd'efd(d)Zd*d+Zd1d'e efd,d-Ze d.ed'ede!e"dffd/d0Z#Z$S)2DeepgramTTSServiceaDeepgram WebSocket-based text-to-speech service. Provides real-time text-to-speech synthesis using Deepgram's WebSocket API. Supports streaming audio generation with interruption handling via the Clear message for conversational AI use cases. _settings)linear16mulawalawNzwss://api.deepgram.comrvoicebase_url sample_rateencodingsettingsapi_keyr!r"r#r$r%c s||jvrtd|dd|jd|jdddd}|dur.|dd||_||_|dur7||t j d |d d d d |d |||_ ||_ ||_ d|_dS) aInitialize the Deepgram WebSocket TTS service. Args: api_key: Deepgram API key for authentication. voice: Voice model to use for synthesis. .. deprecated:: 0.0.105 Use ``settings=DeepgramTTSService.Settings(voice=...)`` instead. base_url: WebSocket base URL for Deepgram API. Defaults to "wss://api.deepgram.com". sample_rate: Audio sample rate in Hz. If None, uses service default. encoding: Audio encoding format. Defaults to "linear16". Must be one of SUPPORTED_ENCODINGS. settings: Runtime-updatable settings. When provided alongside deprecated parameters, ``settings`` values take precedence. **kwargs: Additional arguments passed to parent InterruptibleTTSService class. Raises: ValueError: If encoding is not in SUPPORTED_ENCODINGS. zUnsupported encoding 'z'. Must be one of z, z for WebSocket TTS.Naura-2-helena-enmodelr!languager!TF)r#pause_frame_processingpush_stop_framespush_start_frameappend_trailing_spacer%r)lowerSUPPORTED_ENCODINGS ValueErrorjoinSettings"_warn_init_param_moved_to_settingsr)r! apply_updatesuper__init___api_key _base_url _encoding _receive_task) selfr&r!r"r#r$r%kwargsdefault_settings __class__rrr7@s:    zDeepgramTTSService.__init__returncCdS)zCheck if the service can generate metrics. Returns: True, as Deepgram WebSocket TTS service supports metrics generation. Trr<rrrcan_generate_metricsz'DeepgramTTSService.can_generate_metricsframec&t|IdH|IdHdS)zStart the Deepgram WebSocket TTS service. Args: frame: The start frame containing initialization parameters. N)r6start_connectr<rFr?rrrHzDeepgramTTSService.startcrG)zbStop the Deepgram WebSocket TTS service. Args: frame: The end frame. N)r6stop _disconnectrJr?rrrLrKzDeepgramTTSService.stopcrG)zgCancel the Deepgram WebSocket TTS service. Args: frame: The cancel frame. N)r6cancelrMrJr?rrrNrKzDeepgramTTSService.cancelcsLtIdH|IdH|jr"|js$|||j|_dSdSdS)z5Connect to Deepgram WebSocket and start receive task.N)r6rI_connect_websocket _websocketr; create_task_receive_task_handler _report_errorrCr?rrrIs  zDeepgramTTSService._connectcsBtIdH|jr||jIdHd|_|IdHdS)z6Disconnect from Deepgram WebSocket and clean up tasks.N)r6rMr; cancel_task_disconnect_websocketrCr?rrrMs zDeepgramTTSService._disconnectdeltacsTt|IdH}d|vr|jj|j_||r(|IdH|IdH|S)zApply a settings delta. Args: delta: A :class:`TTSSettings` (or ``DeepgramTTSService.Settings``) delta. Returns: Dict mapping changed field names to their previous values. Nr!)r6_update_settingsrr!r)_sync_model_name_to_metricsrMrI)r<rVchangedr?rrrWs  z#DeepgramTTSService._update_settingsc sVzo|jr|jjtjurWdStdg}|d|jj|d|j |d|j |j dd |}dd |j i}t||d IdH|_d d |jjjD}t|d |d|dIdHWdSty}z.t|d||t|d|dIdHd|_|d|IdHWYd}~dSd}~ww)z;Connect to Deepgram WebSocket API with configured settings.Nz Connecting to Deepgram WebSocketzmodel=z encoding=z sample_rate=z /v1/speak?& AuthorizationToken )additional_headerscSs i|] \}}|dr||qS)zdg-) startswith).0kvrrr s z9DeepgramTTSService._connect_websocket..z0: Websocket connection initialized: {"headers": } on_connected exception:  error: erroron_connection_error)rPstaterOPENrdebugappendrr!r:r#r9r2r8websocket_connectresponseheadersitems_call_event_handler Exceptionrh push_errorr )r<paramsurlrperrrrOs0   "z%DeepgramTTSService._connect_websocketc szpz'|IdH|jr(td|jtddiIdH|jIdHWn,tyU}z t |d|| t |d|dIdHWYd}~nd}~wwWd|_| dIdHdSWd|_| dIdHdSd|_| dIdHw) z+Close WebSocket connection and reset state.Nz%Disconnecting from Deepgram WebSockettypeClosererfrgon_disconnected) stop_all_metricsrPrrlsendjsondumpsclosersrhrtr rr)r<rwrrrrUs* ,z(DeepgramTTSService._disconnect_websocketcCs|jr|jStd)z3Get active websocket connection or raise exception.zWebsocket not connected)rPrsrCrrr_get_websocketsz!DeepgramTTSService._get_websocket context_idc sx|IdH|jr:z|jtddiIdHWdSty9}zt|d|WYd}~dSd}~wwdS)a9Send Clear message to Deepgram when an audio context is interrupted. The Clear message will clear Deepgram's internal text buffer and stop sending audio, allowing for a new response to be generated. Args: context_id: The ID of the audio context that was interrupted. NrxClearz error sending Clear message: )r{rPr|r}r~rsrrh)r<rrwrrron_audio_context_interrupted s ""z/DeepgramTTSService.on_audio_context_interruptedc sR|2z3dHW}t|tr'|}t||jd|d}|||IdHqt|trzft |}| d}|dkrDt d|nM|dkrit d||}||t |dIdH||IdHn(|d krvt d |n|d krt |d | d dnt d|Wqtjyt d|Yqwq6dS)z5Receive and process messages from Deepgram WebSocket.N)rrxMetadatazReceived metadata: FlushedzReceived Flushed: ClearedzReceived Cleared: Warningz warning: descriptionzUnknown warningzReceived unknown message type: zInvalid JSON message: )r isinstancebytesget_active_audio_context_idr r#append_to_audio_contextstrr}loadsgetrtracer remove_audio_contextwarningrlJSONDecodeErrorrh)r<messagectx_idrFmsgmsg_typerrr_receive_messagess@      z$DeepgramTTSService._receive_messagesc sn|jr5zddi}|jt|IdHWdSty4}zt|d|WYd}~dSd}~wwdS)zFlush any pending audio synthesis by sending Flush command. This should be called when the LLM finishes a complete response to force generation of audio from Deepgram's internal text buffer. rxFlushNz error sending Flush message: )rPr|r}r~rsrrh)r<r flush_msgrwrrr flush_audio=s"zDeepgramTTSService.flush_audiotextc Cst|d|dz)|jr|jjtjur|IdHd|d}|t |IdHdVWdSt yQ}zt d|dVWYd}~dSd}~ww)a9Generate speech from text using Deepgram's WebSocket TTS API. 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, plus start/stop frames. : Generating TTS []NSpeak)rxrzUnknown error occurred: rg) rrlrPrjrCLOSEDrIrr|r}r~rsr )r<rr speak_msgrwrrrrun_ttsJs    zDeepgramTTSService.run_tts)N)%rrrrrr3__annotations__r0rrintr7boolrDr rHrrLrrNrIrMrdictrrWrOrUrrrrrrr r __classcell__rrr?rr3sL F    $ (rcseZdZUdZeZeed<dddddddedeed e j d ed ee d ed eeffddZ de fddZedededeedffddZZS)DeepgramHttpTTSServicezDeepgram HTTP text-to-speech service. Provides text-to-speech synthesis using Deepgram's HTTP TTS API. Supports various voice models and audio encoding formats with configurable sample rates and quality settings. rNzhttps://api.deepgram.comrr r&r!aiohttp_sessionr"r#r$r%c  sz|jdddd} |dur|dd|| _|| _|dur!| |tjd|dd| d|||_||_||_ ||_ dS)agInitialize the Deepgram TTS service. Args: api_key: Deepgram API key for authentication. voice: Voice model to use for synthesis. .. deprecated:: 0.0.105 Use ``settings=DeepgramHttpTTSService.Settings(voice=...)`` instead. aiohttp_session: Shared aiohttp session for HTTP requests with connection pooling. base_url: Custom base URL for Deepgram API. Defaults to "https://api.deepgram.com". sample_rate: Audio sample rate in Hz. If None, uses service default. encoding: Audio encoding format. Defaults to "linear16". settings: Runtime-updatable settings. When provided alongside deprecated parameters, ``settings`` values take precedence. **kwargs: Additional arguments passed to parent TTSService class. Nr'r(r!T)r#r-r,r%r) r3r4r)r!r5r6r7r8_sessionr9r:) r<r&r!rr"r#r$r%r=r>r?rrr7ts.   zDeepgramHttpTTSService.__init__rAcCrB)zCheck if the service can generate metrics. Returns: True, as Deepgram TTS service supports metrics generation. TrrCrrrrDrEz+DeepgramHttpTTSService.can_generate_metricsrrc Cst|d|d|jd}d|jdd}|jj|j|jdd}d |i}zy|Id H|j j ||||d 4Id HV}|j d krX| Id H}t d |j d|||Id H|j} d} |j| 2z3d HW} | r||Id Hd} | rt| |jd|dVqk6Wd Id HWd S1Id HswYWd St y} ztdt| VWYd } ~ d Sd } ~ ww)a/Generate speech from text using Deepgram's TTS API. 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, plus start/stop frames. rrz /v1/speakr\zapplication/json)r[z Content-Typenone)r)r$r# containerrN)rpr}ruzHTTP z: TFr)audior# num_channelsrzError getting audio: )rrlr9r8rr!r:r#start_ttfb_metricsrpoststatusrrsstart_tts_usage_metrics chunk_sizecontent iter_chunkedstop_ttfb_metricsr r r) r<rrrvrprupayloadro error_text CHUNK_SIZE first_chunkchunkrwrrrrsR  2"zDeepgramHttpTTSService.run_tts)rrrrrr3rrraiohttp ClientSessionrr7rrDrrr rrrrr?rris6  =(r)&rr} dataclassesrtypingrrrrlogururpipecat.frames.framesrrr r r r r pipecat.services.settingsrpipecat.services.tts_servicerr(pipecat.utils.tracing.service_decoratorsrwebsockets.asyncio.clientrrnwebsockets.protocolrModuleNotFoundErrorrwrhrsrrrrrrrs6  $   8