Pmj| UdZddlmZddlZddlZddlZddlZddlZddlm Z m Z m Z ddddddZ ddd d d d d d d d d d d Z ehdZehdZdZdNdZdOdZdOdZddlmZmZmZmZmZejeZdPdZdQdZdRdSd#Z da!ej"Z#ej"Z$d$a%d$a&d%a'de(d&<da)de(d'<ej*Z+e+,da-d(e(d)<dTd+Z.dQd,Z/da0d-e(d.<da1d-e(d/<da2d0e(d1<da3d-e(d2<da4d3Z5dUd4Z6dVd6Z7 dWdXdDZ8dYdZdFZ9dQdGZ:dUdHZ;d[d\dLZcdlzc-bcfgrrct|tsdS|d}t|tsdS|dS)aShape-safe ``cfg.voice.record_key`` lookup. ``load_config()`` deep-merges raw YAML and preserves scalar overrides, so a hand-edited ``voice: true`` / ``voice: cmd+b`` leaves ``cfg["voice"]`` as a bool/str instead of a dict, and the naive ``.get("voice", {}).get("record_key")`` chain raises AttributeError before voice can even start (Copilot round-11 on #19835). Return ``None`` for malformed shapes so call sites can feed the result straight into the normalizer/formatter and get the documented default. Nvoice record_key) isinstancedictget)rrs 6/home/thesage/.hermes/hermes-agent/hermes_cli/voice.pyvoice_record_key_from_configr%WsU c4 t GGG  E eT " "t 99\ " ""rawstrct|tstS|}|stSd|dD}|stSt |dkrtSt |dkrtS|\}}|dvrtSt|}|stSt |dkrB|dkr|tvrtS|dkr tj dkr|tvrtS||St|}|stS||S) uCoerce ``voice.record_key`` into prompt_toolkit's ``c-x`` / ``a-x`` format. Mirrors the TUI parser contract (``ui-tui/src/lib/platform.ts``) so one config value binds the same shortcut in both runtimes: * non-string / empty / typo'd / bare-char / multi-modifier / reserved ``ctrl+c|d|l`` → documented default ``c-b`` * single-char keys: ``ctrl+o`` → ``c-o`` * named keys: ``ctrl+space`` → ``c-space`` (aliases collapse: ``ctrl+return`` → ``c-enter``) * ``super`` / ``win`` / ``windows`` → ``c-b`` (TUI-only modifiers — prompt_toolkit has no super mod; the CLI binding site is expected to warn when this fallback fires so users see the cross-runtime split, Copilot round-11 on #19835) c^g|]*}||+S)strip).0ps r$ zAnormalize_voice_record_key_for_prompt_toolkit..s- @ @ @1aggii @QWWYY @ @ @r&+>winsuperwindowsrrdarwin)r!r(_DEFAULT_PT_KEYr,lowersplitlen_VOICE_MOD_ALIASESr#_VOICE_RESERVED_CTRL_CHARSsysplatform_VOICE_RESERVED_ALT_CHARS_MAC_VOICE_NAMED_KEYS)r'loweredpartsmodifier_token key_tokennormalized_modnameds r$-normalize_voice_record_key_for_prompt_toolkitrGms{ c3  iikk!!G  @ @ c 2 2 @ @ @E   5zzA~~  5zzQ %NI444'++N;;N  9~~ T ! !i3M&M&M" " d " " ((:::" " -)---  ! !) , ,E  %e % %%r&c t|}|dr d|dd}}nj|dr d|dd}}nHd|vrB|dd\}}|d |ddzdz}nd S|s|dSt |dkr||zS||d z|ddzS) aFRender ``voice.record_key`` for ``/voice status`` in CLI-friendly form. Mirrors the TUI's ``formatVoiceRecordKey``: returns ``Ctrl+B`` / ``Alt+Space`` / ``Ctrl+Enter``. Malformed configs surface as the documented default so status never advertises a shortcut that won't bind (Copilot round-10 on #19835). rzCtrl+r1NrzAlt+r0r2rzCtrl+B)rG startswithr9upperrstripr:)r' normalizedprefixkeymods r$"format_voice_record_key_for_statusrPs?sCCJT"" z!""~   t $ $jn  ##C++SQ#abb')C/x "}}S!!! 3xx1}} ## CFLLNN "SW ,,r&)create_audio_recorderis_voice_stop_phraseis_whisper_hallucinationplay_audio_filetranscribe_recordingmsgNonectjdddkrdS t d|t jddS#ttf$rYdSwxYw)u^Emit a debug breadcrumb when HERMES_VOICE_DEBUG=1. Goes to stderr so the TUI gateway wraps it as a gateway.stderr event, which createGatewayEventHandler shows as an Activity line — exactly what we need to diagnose "why didn't the loop auto-restart?" in the user's real terminal without shipping a separate debug RPC. Any OSError / BrokenPipeError is swallowed because this fires from background threads (silence callback, TTS daemon, beep) where a broken stderr pipe must not kill the whole gateway — the main command pipe (stdin+stdout) is what actually matters. HERMES_VOICE_DEBUG1Nz[voice] T)fileflush) osenvironr#r,printr=stderrBrokenPipeErrorOSError)rVs r$_debugrds z~~*B//55773>>  SZt<<<<<< W %     sAA0/A0boolc ddlm}ddlm}|di}t |t r!||dddSn#t$rYnwxYwdS)z=CLI parity: voice.beep_enabled in config.yaml (default True).r) load_config)is_truthy_valuer beep_enabledT)default)hermes_cli.configrgutilsrhr#r!r" Exception)rgrh voice_cfgs r$_beeps_enabledros  111111))))))KMM%%gr22 i & & V#?9==#F#FPTUUU U V      4sAA## A0/A0r2 frequencyintcountctsdS ddlm}|||dS#t$r }t d|d|Yd}~dSd}~wwxYw)uUAudible cue matching cli.py's record/stop beeps. 880 Hz single-beep on start (cli.py:_voice_start_recording line 7532), 660 Hz double-beep on stop (cli.py:_voice_stop_and_transcribe line 7585). Best-effort — sounddevice failures are silently swallowed so the voice loop never breaks because a speaker was unavailable. Nr) play_beeprprrzbeep z Hz failed: )rotools.voice_modertrmrd)rprrrtes r$ _play_beeprx s   2...... IU333333 2220y00Q001111111112s' AA  AFT_continuous_auto_restart_continuous_recorderOptional[Callable[[], bool]]_voice_busy_probeprobec |adS)u$Register a callable that returns True while the agent is mid-turn. Called by the hosting surface (tui_gateway registers one that checks every session's ``running`` flag). ``None`` clears it. The probe must be cheap and thread-safe — it runs on the silence-callback thread. N)r|r}s r$set_voice_busy_prober;sr&ctsdSt}|dS t|S#t$rYdSwxYw)a0True while silent cycles must NOT count toward the no-speech limit. Held when TTS is playing (the user is listening) or when the registered busy probe reports the agent mid-turn (the user is waiting). Fail-open to "not held" so a broken probe can never make the voice chat immortal. TNF) _tts_playingis_setr|rermrs r$_voice_activity_heldrFse    t E }uEEGG}} uus? A  A Optional[Callable[[str], None]]_continuous_on_transcript_continuous_on_statusOptional[Callable[[], None]]_continuous_on_silent_limit_continuous_on_stop_phrasect5t$ttddr ddddSt}||addddS#1swxYwYdS)uBegin capturing from the default input device (push-to-talk). Idempotent — calling again while a recording is in progress is a no-op. N is_recordingF)_recorder_lock _recordergetattrrQstart)recs r$start_recordingris   WY%N%N $%%  sA%$A%%A),A) Optional[str]cjt5t}dadddn #1swxYwY|dS|}|sdS t|}nz#t$rm}t d|Yd}~ tj |rtj |dSdS#t$rYdSwxYwd}~wwxYw tj |rtj |nZ#t$rYnNwxYw# tj |rtj |ww#t$rYwwxYwxYw| dsdS| dpd }|rt|rdS|S)zStop the active push-to-talk recording, transcribe, return text. Returns ``None`` when no recording is active, when the microphone captured no speech, or when Whisper returned a known hallucination. Nzvoice transcription failed: %ssuccess transcriptrZ)rrstoprUrmloggerwarningr^pathisfileunlinkr#r,rS)rwav_pathresultrwtexts r$stop_and_transcriberxsD   {txxzzH t %h// 7;;;ttt w~~h'' $ (##### $ $    DD    w~~h'' $ (###    D  w~~h'' $ (#### $    D  ::i t JJ| $ $ * 1 1 3 3D +D11t Ks ""AD C  C;D3B77 CCC  D3D DDE3E  E EEEE@ on_transcriptCallable[[str], None] on_statuson_silent_limitsilence_thresholdsilence_durationfloat auto_restartmax_recording_secondson_stop_phrasect5trtd ddddStrtd ddddSda|a|a|a|a|a|rda tta |t_ |t_ t|tt frt|t"s|dkr|ndt_t}dddn #1swxYwYtd|d |d t'd d  |t*nv#t,$ri} t.d| tdt3| jd| t5dadddn #1swxYwYd} ~ wwxYw|r |dn#t,$rYnwxYwdS)uBStart a VAD-driven continuous recording loop. The loop calls ``on_transcript(text)`` each time speech is detected and transcribed successfully. If ``auto_restart`` is True, it auto-restarts for the next turn and resets the no-speech counter for that loop. If ``auto_restart`` is False, the first silence-triggered transcription ends the loop and reports ``"idle"``; no-speech counts are retained across starts so a push-to-talk caller can still enforce the three-strikes guard. After ``_CONTINUOUS_NO_SPEECH_LIMIT`` consecutive silent cycles (no speech picked up at all) the loop stops itself and calls ``on_silent_limit`` so the UI can reflect "voice off". Returns False if a previous stop is still transcribing/cleaning up; otherwise returns True. Idempotent — calling while already active is a successful no-op. ``on_status`` is called with ``"listening"`` / ``"transcribing"`` / ``"idle"`` so the UI can show a live indicator. ``max_recording_seconds`` is the hard cap on a single recording's length (``voice.max_recording_seconds``); any non-positive or non-numeric value disables the cap, preserving the previous unbounded behaviour. ``on_stop_phrase`` is called with the (stripped) transcript when the user utters a bare voice stop phrase (``voice.stop_phrases``, default "stop"). The loop halts first, so the consumer only needs to reflect "voice off" — exactly like the user pressing the manual stop control. When omitted, ``on_silent_limit`` fires instead so legacy callers still turn voice off. u*start_continuous: already active — no-opNTu6start_continuous: stop/transcribe in progress — busyFrrz#start_continuous: begin (threshold=z , duration=zs)pr2ruon_silence_stopz(failed to start continuous recording: %sz#start_continuous: rec.start raised : listening)_continuous_lock_continuous_activerd_continuous_stoppingryrrrr_continuous_no_speech_countrzrQ_silence_threshold_silence_durationr!rqrre_max_recording_secondsrxr_continuous_on_silencermrerrortype__name__) rrrrrrrrrrws r$start_continuousrsT ##   ? @ @ @########   K L L L ########"#/ $1! )&5#%3"  ,*+ '  '#8#:#: 2C/1A. /#u>> 4d;; &)) " ! 3#=###############@ `.?``L\```A&&&& "8 9999  ?CCCLT!WW5ELLLLMMM  ' '!&  ' ' ' ' ' ' ' ' ' ' ' ' ' ' '   Ik " " " "    D  4slC&C&B C&&C*-C*D55 F(?A F#F F#F F#F F##F(. F:: GGforce_transcribec  t5ts ddddSdat}tt t t t}|o| |duadadadada sda dddn #1swxYwY|1|r rr dn#t$rYnwxYw | n|#t$ro}t d| |n2#t$r%}t d|Yd}~nd}~wwxYwd Yd}~nd}~wwxYw fd}tj|d dS |n2#t$r%}t d|Yd}~nd}~wwxYwt5dadddn #1swxYwYt%d d r d dS#t$rYdSwxYwdS)u?Stop the active continuous loop and release the microphone. Idempotent — calling while not active is a no-op. If ``force_transcribe`` is True, the recorder stops synchronously, then transcription/cleanup runs on a background thread before reporting ``"idle"``. Otherwise the buffer is discarded. NFr transcribingzfailed to stop recorder: %szfailed to cancel recorder: %scd}d} r t }|dr<|dpd}|rt|s|}tj rt j n9#tj rt j wwxYwn2#t$r%}t d|Yd}~nd}~wwxYwt|ot|}|rEtd|d|pd}d}  |n  n#t$rYnwxYw|r?  |n2#t$r%}t d|Yd}~nd}~wwxYw rt}t5|s|rd an0|rtd nt d z at t"k}|rd adddn #1swxYwY|rr n#t$rYnwxYwt%d d t5dadddn #1swxYwY r  ddS#t$rYdSwxYwdS#t|ot|}|rEtd|d|pd}d}  |n  n#t$rYnwxYw|r?  |n2#t$r%}t d|Yd}~nd}~wwxYw rt}t5|s|rd an0|rtd nt d z at t"k}|rd adddn #1swxYwY|rr n#t$rYnwxYwt%d d t5dadddn #1swxYwY r  dw#t$rYwwxYwwxYw)NFrrrZz&failed to stop/transcribe recorder: %szstop_continuous: stop phrase u — ending voice chat!on_transcript callback raised: %srzAstop_continuous: silent cycle ignored (agent busy or TTS playing)r2r1ruidle)rUr#r,rSr^rrrrmrrrerRrdrrr_CONTINUOUS_NO_SPEECH_LIMITrxr)r should_haltrrrw stop_phrase stop_textheldrrrrtrack_no_speechrs r$_transcribe_and_cleanupz0stop_continuous.._transcribe_and_cleanup;sD,0 # J! 44%9(%C%CF%zz)446(. <(@(@(FB'M'M'O'O#'!60H0N0N!615J!w~~h774 " ( 3 3 3 "w~~h774 " ( 3 3 3 34 PPPNN#KQOOOOOOOOP#'z'V6J:6V6V"W"WK"!`J```%/$4" %) !-9 .y 9 9 9 9!0!< / 1 1 1(!!! D!!SS)M*5555(SSS"NN+NPQRRRRRRRRS'%355-DD)D[D>? ; ;!%D!'%B!"!"!"!" !? ; ;!%D!'%B!"!"!"!" !> F-F((F-8H  HH H&& H32H3II!$I!+ I88 JJ :P K! P ! K.+P -K..P 4 L?P  L/ L*%P *L//P  8N  P  N P N P  N('P ( N52P 4N55P O P O# #P &O# 'P - O98P 9 PP PP T)targetdaemonrr1rur)rrrzrrrrryrrrmrrrcancel threadingThreadrrx) rrrrw cancel_errorrrrrrrrs @@@@@@r$stop_continuousr s ,,!  ,,,,,,,,#") 1 53/ *?BBBBBBBB C %%$%%%%%%%%%%%%%%% A&&&&  If         DD s A1AA11A58A5 B BB#B88 D1D,C32D,3 D"=DD,D""D,,D1)E>> F-F((F-7GG  G % G22 H?HcRt5tcdddS#1swxYwYdS)z5Whether a continuous voice loop is currently running.N)rrr+r&r$is_continuous_activerso ""!""""""""""""""""""s   c tdt5tstd ddddSt}t}t }t }t}dddn #1swxYwY|tddS|r |dn#t$rYnwxYw| }t|dd}td|d |d td d d}|r t|}t|d} |dpd} |d} td| d| d| | r| rt!| s| }nY#t$rL} t"d| tdt'| jd| Yd} ~ nd} ~ wwxYw t*j|rt+j|nZ#t$rYnNwxYw# t*j|rt+j|ww#t$rYwwxYwxYwt|ot3|} | r|pdnd}| rtd|dd}|duo| o t5}t5tstd ddddS|rdan|rtdn | s t6dz a| pt6t8k}t6}dddn #1swxYwY|rA|r? ||n2#t$r%} t"d| Yd} ~ nd} ~ wwxYw|rtd| rd n|d!d t5d"adadddn #1swxYwY| r| ||n.#t$rYn"wxYw|r |n#t$rYnwxYw |n#t$rYnwxYw|r |d#n#t$rYnwxYwdSt<std$t< d%&ddl!}|"d't5tstd( ddddS dddn #1swxYwYtFrtd)|d td*d  |$tJ+n#t$r} t"&d,| td-t'| jd| t5d"adddn #1swxYwY|r |d#n#t$rYnwxYwYd} ~ dSd} ~ wwxYw|r |d.dS#t$rYdSwxYwdStd/t5d"adddn #1swxYwY|r |d#dS#t$rYdSwxYwdS)0u AudioRecorder silence callback — runs in a daemon thread. Stops the current capture, transcribes, delivers the text via ``on_transcript``, and — if the loop is still active — starts the next capture. Three consecutive silent cycles end the loop. z_continuous_on_silence: firedu/_continuous_on_silence: loop inactive — abortNu-_continuous_on_silence: no recorder — abortr _peak_rmsz$_continuous_on_silence: rec.stop -> z (peak_rms=)rr1rurrrZrz._continuous_on_silence: transcribe -> success=z text=z err=z#continuous transcription failed: %sz*_continuous_on_silence: transcribe raised rz$_continuous_on_silence: stop phrase u — ending loopu@_continuous_on_silence: stopped during transcribe — no restartrzH_continuous_on_silence: silent cycle ignored (agent busy or TTS playing)r2rz!_continuous_on_silence: halting (z stop phrasez silent cyclesFrz1_continuous_on_silence: waiting for TTS to finish<)timeout333333?z5_continuous_on_silence: stopped while waiting for TTSz3_continuous_on_silence: restarting loop (no_speech=rrz*failed to restart continuous recording: %sz'_continuous_on_silence: restart raised rz9_continuous_on_silence: auto_restart=False, stopping loop)'rdrrrzrrrrrmrrrxrUrer#r,rSrrrrr^rrrrRrrrrrrwaittimesleepryrrr)rrrrrrpeak_rmsrrrrerrrwrr _silence_heldr no_speech_times r$rrs  *+++ 44!  D E E E 44444444#1 ) 53444444444444444 {>???  In % % % %    D xxzzHsK,,H QxQQhQQQ A&&&& $J )(33F 6::i0011GJJ|,,299;;D**W%%C -----%(--    "4 "(@(F(F "!  Y Y Y NN@! D D D WQ@PWWTUWW X X X X X X X X Y 7>>(++(Ih'''     7>>(++(Ih''''(    zF&::&F&FGGK&19!rrI TjTTTUUU  4'0 O0-// 00!  U V V V  00000000  -*+ ' '  - .     - '1 , '! '+F F 0 #000000000000000&CmC C M* % % % % C C C NN> B B B B B B B B C R!,N Y2N2N2N R R R    , ,!& *+ ' , , , , , , , , , , , , , , ,  >5 y))))       !!!!      JJLLLL    D     &!!!!        BCCC"%%% C  % NOOO                          QYQQQRRRS****  II& )r output_path file_pathzspeak_text: playing z (z bytes).r2z.oggz*speak_text: TTS tool produced no audio at zVoice TTS playback failed: %szspeak_text raised r)3r,retempfilerrrrzrrrmrrrclearrdrrtools.tts_streamingrrrsetrrrtools.tts_text_normalizerr:sub MULTILINEr^makedirsrjoin gettempdirstrftimer!r(jsonloadsrgetsizer#rTlistrsplitaddrrcrr)rrrrrpaused_recordingrwrrrrtts_textmp3_path raw_result tts_result audio_path cleanup_pathsrogg_paths r$ speak_textrs  tzz|| IIIOOOKKK   J J  J$0,neDD1 J$++---#'   J J JEqIIIIIIII J J J J J J J J J J J J J J J J I6F I I IJJJf666666 Y F F F F F F 7 7 7 7 7 7))*:*:*<*<==I(z::J %&&&   JJsOOO!  % *>*J,22,B3HIIII$Eq                    U Y Y Y W!WWW X X X X X X X X Y  ( D D D D D D**44@@@HH ( ( (&)$ii$&6&6tETE{{DHvv/h??Hvv6xHHHvvor8<*J,22,B3HIIII$Eq                    ] BGLL!4!4!6!6GGRVWWWW7<<    ! !  74==11 7 7 7   N3x==NNHNNOOO((hHMMM  3=j#3N3NVJ///TVJJ   JJJ   w~~h'' A27??8+D+D+I+I# 44@J 7>>* % % N"'//**E*E*I*I \*\\ 8S8S\\\ ] ] ] J ' ' ' !+X 6  //00D#{{32215>H!%%h////)((Dw~~d++( $(     L LL M M M ===6:::;DGG$4;;;;<<<<<<<<= %&&&   JJsOOO!  % *>*J,22,B3HIIII$Eq                     %&&&   JJsOOO!  % *>*J,22,B3HIIII$Eq                 s%B.A0/B.0 B:BB.BB..B25B2&[-/G2#G"3/F#"G"# G-G G" GG""G&)G&1[2 H<H[H[H20[2D2M'$[&M''[3P2/O32P23 P"=PP2P""P22P69P6B?[+T-,[- T<9[;T<b; ?c)rrrr)r'rrr()rVr(rrW)rre)r2)rprqrrrqrrW)r}r{rrW)rrW)rr)NNrrTrN)rrrrrrrrqrrrrerrrrrre)F)rrerrW)N)rr(rrrre)rr(rrrrW)>__doc__ __future__rrloggingr^r=rtypingrrrr;r@ frozensetr<r?r7r%rGrPrvrQrRrSrTrU getLoggerrrrdrorxrLockrrrrry__annotations__rzrrrr|rrrrrrrrrrrrrrrrr+r&r$r&s*#"""""  **********             &'Y77!* /// : :####,J&J&J&J&Z---->  8 $ $    *     22222$  !!"9>##!%%%%%  y  376666&>BAAAA9=====<@@@@@ ?CBBBB    ''''^2648 !#&6:bbbbbJWWWWWt"""" DDDDTBOOOOOOOr&