qi/:(dZddlZddlZddlmZmZddlZddlm Z ddl m Z m Z m Z ddlmZmZmZddlmZmZ ddlZGd d eZGd deZy#e$r7Ze j0dee j0d ed edZ[wwxYw)aKrisp turn analyzer for end-of-turn detection using Krisp VIVA SDK. This module provides a turn analyzer implementation using Krisp's turn detection (Tt) API to determine when a user has finished speaking in a conversation. Note: This analyzer uses a different model than KrispVivaFilter. The model path can be specified via the KRISP_VIVA_TURN_MODEL_PATH environment variable or passed directly to the constructor. N)OptionalTuple)logger)KrispVivaSDKManagerint_to_krisp_frame_durationint_to_krisp_sample_rate)BaseTurnAnalyzerBaseTurnParamsEndOfTurnState) MetricsDataTurnMetricsDataz Exception: z?In order to use KrispVivaTurn, you need to install krisp_audio.zMissing module: c.eZdZUdZdZeed<dZeed<y)KrispTurnParamsaeConfiguration parameters for Krisp turn analysis. Parameters: threshold: Probability threshold for turn completion (0.0 to 1.0). Higher values require more confidence before marking turn as complete. frame_duration_ms: Frame duration in milliseconds for turn detection. Supported values: 10, 15, 20, 30, 32. g? thresholdframe_duration_msN) __name__ __module__ __qualname____doc__rfloat__annotations__rintT/opt/pipecat/venv/lib/python3.12/site-packages/pipecat/audio/turn/krisp_viva_turn.pyrr(sIusrrc eZdZdZddddddeedeedeeded df fd Zd Z defd Z deffd Z e d e fdZe d eefdZe d efdZe d efdZdeded efdZd eeeeffdZdZxZS) KrispVivaTurnzTurn analyzer using Krisp VIVA SDK for end-of-turn detection. Uses Krisp's turn detection (Tt) API to determine when a user has finished speaking. This analyzer requires a valid Krisp model file to operate. N) model_path sample_rateparamsapi_keyr r!r"r#returnc$t|| tj|d|_ |xstjd|_ |js tjdtd |jjd s t d tjj|jst!d |j|xs t#|_d|_d|_d|_t-|_d|_d|_g|_t6j8|_d|_d|_d|_ |r|nd } |jC||_y#t $r}d|_t d|d}~wwxYw#t $r5}tjd|dd|_t d||d}~wwxYw#t $r)|jrtjDd|_wxYw)ahInitialize the Krisp turn analyzer. Args: model_path: Path to the Krisp turn detection model file (.kef extension). If None, uses KRISP_VIVA_TURN_MODEL_PATH environment variable. sample_rate: Optional initial sample rate for audio processing. If provided, this will be used as the fixed sample rate. params: Configuration parameters for turn analysis behavior. api_key: Krisp SDK API key. If empty, falls back to the KRISP_VIVA_API_KEY environment variable. Raises: ValueError: If model_path is not provided and KRISP_VIVA_TURN_MODEL_PATH is not set. Exception: If model file doesn't have .kef extension. FileNotFoundError: If model file doesn't exist. RuntimeError: If Krisp SDK initialization fails. )r!)r#TFz Failed to initialize Krisp SDK: NKRISP_VIVA_TURN_MODEL_PATHzEModel path is not provided and KRISP_VIVA_TURN_MODEL_PATH is not set.z.Model path for KrispVivaTurn must be provided.z.kefz%Model is expected with .kef extensionzModel file not found: i>)Failed to create turn detection session: exc_info)#super__init__racquire _sdk_acquired Exception RuntimeErrorosgetenv _model_pathrerror ValueErrorendswithpathisfileFileNotFoundErrorr_params _tt_session_preload_tt_session_samples_per_frame bytearray _audio_buffer_speech_triggered_last_probability_frame_probabilitiesr INCOMPLETE _last_state_speech_stopped_time_e2e_processing_time_ms _last_metrics_create_tt_sessionrelease)selfr r!r"r#epreload_sample_rate __class__s rr+zKrispVivaTurn.__init__=s2 [1 G  ' ' 8!%D  - )TRYY7S-TD ## [!!QRR##,,V4 GHH77>>$"2"23'*@AQAQ@R(STT!6_%6DL#D '+D $&*D #!*D &+D "%)D "(*D %-88D 9=D %<@D (<@D 2=+%  [+/+B+BCV+W(M G!&D !A!EF F GN [ HLW[\+/("%Nqc#RSYZZ [  !!#++-%*"   sAE6D-GF6 F?FF G%0GGG2Hc:K|jr[ t|dr|jd|_t|dr|jd|_t j d|_yy#t $r$}tjd|dYd}~yd}~wwxYww)z1Release SDK reference when analyzer is destroyed.r:Nr;FzError in __del__: Tr() r-hasattrr:r;rrHr.rr3)rIrJs rcleanupzKrispVivaTurn.cleanups    F4/D4D4D4P'+D$4!67DC C8 )C33C8c|j|k(ryt| | |j|j|_|j y#t $r+}tjd|dd|_Yd}~yd}~wwxYw)zSet the sample rate and create/update the turn detection session. Args: sample_rate: The sample rate to set. Nr'Tr() _sample_rater*set_sample_raterGr:clearr.rr3)rIr!rJrLs rr_zKrispVivaTurn.set_sample_rates|    +   , $#66t7H7HID  JJL $ LLDQCHSW X#D   $s0A B!BBc|jS)zGet all probabilities from the last append_audio call. Returns: List of probability values for each frame processed in the last append_audio call. )rArIs rframe_probabilitiesz!KrispVivaTurn.frame_probabilitiess(((rc|jS)zGet the last turn probability value computed. Returns: Last probability value, or None if no frames have been processed yet. )r@rbs rlast_probabilityzKrispVivaTurn.last_probability%%%rc|jS)zCheck if speech has been detected and triggered analysis. Returns: True if speech has been detected and turn analysis is active. )r?rbs rspeech_triggeredzKrispVivaTurn.speech_triggeredrfrc|jS)zGet the current turn analyzer parameters. Returns: Current turn analyzer configuration parameters. )r9rbs rr"zKrispVivaTurn.paramss||rbuffer is_speechc|j:tjdtj|_tjS|j :tjdtj|_tjS |jj|g|_ t|jdz}||j z}|dk(r%tj|_tjS||j z}|dz}t|jd|}|j|d|_tj|tj}|jtj dz } | j#d|j } tj} | D]m} |r7|j$stj&d d|_d |_d|_n1|j$r%|j*t-j.|_|jj1| j3} | dkr| |_|jj7| |j$s| |j8j:k\s|j*\t-j.|j*z d z|_t=d d | |j( |_tj@dtjB} |jEn| |_| S#tF$r=}tjHd|d tj}||_|cYd}~Sd}~wwxYw)aAppend audio data for turn analysis. Args: buffer: Raw audio data bytes to append for analysis. is_speech: Whether the audio buffer contains detected speech. Returns: Current end-of-turn state after processing the audio. NzextendrAlenbytesnp frombufferint16astypefloat32reshaper?tracerErDtime perf_counterprocesstolistr@appendr9rr rFdebugCOMPLETEr`r.r3)rIrjrk total_samplesnum_complete_framescomplete_samples_countbytes_to_processaudio_to_process audio_int16 audio_float32framesstateframeprobrJ error_states r append_audiozKrispVivaTurn.append_audios    # NNY Z-88D !,, ,  " " * NNT U-88D !,, ,V     % %f -)+D % 2 23q8M"/43J3J"J "a'#1#<#< %000%84;R;R%R "59 $T%7%78I9I%JK !%!3!34D4E!FD --(8IK'..rzz:WDM"**2t/F/FGF"--E . 11 %MN7;4-1D*04D---$2K2K2S484E4E4G1 ''// ?!8*.&))006 ))ddll6L6L.L00< --/$2K2KK 8!4.=&5(,(,373O3O .*  ':<*33EJJL]. b %D L  LL>qcBT R(33K*D    s3A2L0FL0L0BL00 M692M1+M61M6cLK|j}d|_|j|fSw)zAnalyze the current audio state to determine if turn has ended. Returns: Tuple containing the end-of-turn state and optional metrics data. Returns the last state determined by append_audio(). N)rFrC)rImetricss ranalyze_end_of_turnz!KrispVivaTurn.analyze_end_of_turn_s,$$!((s"$c~d|_|jjtj|_d|_y)z-Reset the turn analyzer to its initial state.FN)r?r>r`r rBrCrDrbs rr`zKrispVivaTurn.clearls2!&   ")44$(!r)rrrrrstrrrr+rOrGr_propertylistrcrreboolrhr"rwr rrr rr` __classcell__)rLs@rrr6s0%)%),0 PSMPc] P ) P  P Pd F]c]>$3$$)T))&(5/&&&$&&j5jTjnjX )5+AV1V+W ))rr)rr0rtypingrrnumpyrxlogururpipecat.audio.krisp_instancerrr%pipecat.audio.turn.base_turn_analyzerr r r pipecat.metrics.metricsr r rRModuleNotFoundErrorrJr3r.rrrrrrs " cb@,  n  {)${)),FLL;qc"#FLLRS &qc* ++,sAB2B  B