qiXddlmZddlmZmZddlmZddlm Z ddl m Z GddZ y ) ) annotations) FfiClient FfiHandle)ffi_pb2) get_address) AudioFramecNeZdZdZddddd d dZd dZd dZd dZy) AudioProcessingModulez Provides WebRTC audio processing capabilities including echo cancellation, noise suppression, high-pass filtering, and gain control. F)echo_cancellationnoise_suppressionhigh_pass_filterauto_gain_controlcbtj}||j_||j_||j_||j_tjj|}t|jjjj|_y)a Initialize an AudioProcessingModule instance with the specified audio processing features. Args: echo_cancellation (bool, optional): Whether to enable echo cancellation. noise_suppression (bool, optional): Whether to enable noise suppression. high_pass_filter (bool, optional): Whether to enable a high-pass filter. auto_gain_control (bool, optional): Whether to enable auto gain control. N) proto_ffi FfiRequestnew_apmecho_canceller_enablednoise_suppression_enabledhigh_pass_filter_enabledgain_controller_enabledrinstancerequestrapmhandleid _ffi_handle)selfr r rrreqresps A/opt/pipecat/venv/lib/python3.12/site-packages/livekit/rtc/apm.py__init__zAudioProcessingModule.__init__s{"""$-> *0A -/? ,.? +!!))#.$T\\%5%5%<%<%?%?@ct|jts0t|jtr0|jjrt |j|_t j}|jj|j_ t|j|j_ t|j|j_|j |j_|j"|j_t$j&j)|}|jj*rt-|jj*y)a\ Process the provided audio frame using the configured audio processing features. The input audio frame is modified in-place (if applicable) by the underlying audio processing module (e.g., echo cancellation, noise suppression, etc.). Important: Audio frames must be exactly 10 ms in duration. N) isinstance_databytes memoryviewreadonly bytearrayrrrrapm_process_stream apm_handlerdata_ptrlensize sample_rate num_channelsrrrerror RuntimeErrorrdatarr s r!process_streamz$AudioProcessingModule.process_stream)s djj% ( tzz: .4::3F3F"4::.DJ""$,0,<,<,C,C)*5djj*A'&)$**o#-1-=-=*.2.?.?+!!))#.  " " ( (t66<<= = )r#ct|jts0t|jtr0|jjrt |j|_t j}|jj|j_ t|j|j_ t|j|j_|j |j_|j"|j_t$j&j)|}|j*j,rt/|j*j,y)a Process the reverse audio frame (typically used for echo cancellation in a full-duplex setup). In an echo cancellation scenario, this method is used to process the "far-end" audio prior to mixing or feeding it into the echo canceller. Like `process_stream`, the input audio frame is modified in-place by the underlying processing module. Important: Audio frames must be exactly 10 ms in duration. N)r%r&r'r(r)r*rrrrapm_process_reverse_streamr,rr-r.r/r0r1rrrr+r2r3r4s r!process_reverse_streamz,AudioProcessingModule.process_reverse_streamDs djj% ( tzz: .4::3F3F"4::.DJ""$484D4D4K4K&&12=djj2I&&/.1$**o&&+595E5E&&26:6G6G&&3!!))#.  " " ( (t66<<= = )r#c@tj}|jj|j_||j_tjj|}|jjrt|jjy)a! This must be called if and only if echo processing is enabled. Sets the `delay` in ms between `process_reverse_stream()` receiving a far-end frame and `process_stream()` receiving a near-end frame containing the corresponding echo. On the client-side this can be expressed as delay = (t_render - t_analyze) + (t_process - t_capture) where, - t_analyze is the time a frame is passed to `process_reverse_stream()` and t_render is the time the first sample of the same frame is rendered by the audio hardware. - t_capture is the time the first sample of a frame is captured by the audio hardware and t_process is the time the same frame is passed to `process_stream()`. N) rrrrapm_set_stream_delayr,delay_msrrrr2r3)rr<rr s r!set_stream_delay_msz)AudioProcessingModule.set_stream_delay_ms`s{ ""$.2.>.>.E.E  +,4  )!!))#.  $ $ * *t88>>? ? +r#N) r boolr r>rr>rr>returnNone)r5r r?r@)r<intr?r@)__name__ __module__ __qualname____doc__r"r6r9r=r#r!r r sb#("'!&"' A A A  A  A A4>6>8@r#r N) __future__r _ffi_clientrr_protorr_utilsr audio_framer r rFr#r!rLs""-(#n@n@r#