+ jHj(~Rt^RIHt^RIt^RIt^RIt^RIHt^RIH t ]P!R4t !RR4t R#)zXVoice-clip transcription for inbound messages (moved out of ``gateway/run_inbound.py``).) annotationsN)Optional)_UNSETz gateway.runc]tRt^tRtRRRlltRRltRRRlltR RR R /R R lltR ] /RRllt Rt ] RRl4t ]RRl4tRRltRtR#) GatewayInboundTranscriptionMixinzCSTT enrichment and the once-only echo of pending voice transcripts.Nc(V^8dQhRRRRRRRR/#)  user_textstr audio_paths list[str] clip_memozOptional[dict]returntuple[str, list[str]])formats"F/home/ubuntu/.hermes/hermes-agent/gateway/run_inbound_transcription.py __annotate__-GatewayInboundTranscriptionMixin.__annotate__s,010101+401AO01 01c R"^RIHp\\P V44p\ VP RR4'g.pVFYp\PPV4pV!V4GRjxL pV'dRV R2MRp VPRV V R 24K[ V'd$VPR PV4V4.3#T.3#^R I Hp Hp .p .pTF}pTeYc9d Y6,wppM=\"P'RT4TP)YjT 4GRjxL wppTeTT3Y6&TeTPT4T PT4K T 'd"TPR PT 4T4pY3#EL \ d5p \"P%R T 4TPR T4.3uRp ? #Rp ? ii;iL \*dCp \"P%RT 4T PTP-T44Rp ? EK<Rp ? ii;i5i)umTranscribe voice clips with the configured STT provider and prepend the transcripts → ``(enriched_text, successful_transcripts)``; the transcripts (input order; empty if every clip failed or STT is disabled) let callers echo them back before the agent loop. ``clip_memo`` (path → ``_transcribe_one_clip`` result) skips clips an earlier pass already transcribed.)_probe_audio_duration stt_enabledTNz (duration: )z [The user sent a voice message: ] )transcribe_audiotranscribe_audio_local_fallbackz$Transcription module unavailable: %sz([voice message could not be transcribed]zTranscribing user voice: %szTranscription error: %s) gateway.runrlistdictfromkeysgetattrconfigospathabspathappend_prepend_media_prefixjointools.transcription_toolsrrModuleNotFoundErrorloggererrordebug_transcribe_one_clip Exception_untranscribed_audio_note)selfr r r rnotesr&abs_path duration_strsuffixrreenriched_partssuccessful_transcripts transcriptnotes&&&& r"_enrich_message_with_transcriptionCGatewayInboundTranscriptionMixin._enrich_message_with_transcriptions 64==56 t{{M488E#77??40%:8%DD ;G< ~Q7R ?z&QRST $ RWD..v{{5/A9Mhjj j\ehjj j i ,.D L(T->'0$JLL!>E-1-F-F0O.($J!,+5t*< )*11*=%%d+ $ 226;;~3NPYZI00G E# i LL? C--.XZcdfhh h i( L 6:%%d&D&DT&JKK LsA3H'5F6 H'*H'-'H'F H''AG(G)7G H',%H' G)G GH' GH'G H$"6HH'H$$H'cV^8dQhRR/#)rrr r)rs"rrrEs  9 rc ^RIHp\\VRR4;'g.4UUu.uFwr4V!W4'gKVNK upp#uuppi)z7Return STT-eligible paths from a pending voice message.)_event_media_is_stt_input media_urlsN)rrA enumerater#)r3eventrAir&s&& r_pending_event_audio_paths;GatewayInboundTranscriptionMixin._pending_event_audio_pathsEsM9 )'%t*L*R*RPR S SWQ(2 D S   s A A c V^8dQhRRRR/#)rr z Optional[str]rztuple[str | None, list[str]]r)rs"rrrMs55 -5 %5rc "\VR4'd-VP\\VR.4;'g.43#VP V4pV'g"VeV.3#\VRR4;'gR.3#VeTM\VRR4;'gRp\VRR4pVf /pWQnVP WCV4GRjxL wrgVP V4V8wdMVP V4pVeTM\VRR4;'gRpVP WCV4GRjxL wrgKbWan\V4VnWg3#LL#5i)u Transcribe a pending audio event once and cache the result on the event: the interrupt monitor and the pending-drain path both need it — one STT call and one echo per message. ``user_text`` is a deliberate override of the caption; ``None`` reads ``event.text`` live._gateway_pending_stt_text _gateway_pending_stt_transcriptsNtextr_gateway_pending_stt_clips)hasattrrJr r#rFrMr=rK)r3rDr r rLr enriched_textr:s&&& r$_transcribe_pending_audio_event_onceEGatewayInboundTranscriptionMixin._transcribe_pending_audio_event_onceMsi 55 6 622DHjln9o9u9usu4vv v55e<  ) 59cee eGE6SW:a:a9;5 1M1+8'156L1M.441| 5sA6E  E E 6E 1E EAE E &E '"E  E metadata log_context Transcriptc$V^8dQhRRRRRR/#)r transcriptsr rSr rNoner)rs"rrrls&  3<    rc "V'dVP4'dVfR#\\VR^4;'g^4p\V\ V44VnVP W#WGRWVR7GRjxL R#L5i)u>Echo pending-event STT transcripts to the chat at most once. Tracked as a COUNT (not a set — identical transcripts are distinct deliveries): ``merge_pending_message_event`` can append a second voice note and invalidate the cache; the re-run returns earlier transcripts as a prefix, so only the unsent tail is echoed.N_gateway_pending_stt_echoedrRrS)_should_echo_stt_transcriptsintr#maxlenrY_echo_stt_transcripts)r3rDadaptersourcerVrRrSalready_echoeds&&&&&$$ r"_echo_pending_stt_transcripts_onceCGatewayInboundTranscriptionMixin._echo_pending_stt_transcripts_oncelsy$"C"C"E"E WU,I1MRRQRS,/K@P,Q)(( [9H)   sBB:B:B;Bc$V^8dQhRRRRRR/#)rrLr rSrrr)rs"rrr|s$,/AD rc "VPV4'gV.3#VPV4GRjxL wrxV\Jd!VPW0P V44pVP WW8WeR7GRjxL T;'gTV3#LZL \ d%p \PRYY4T.3uRp ? #Rp ? ii;i5i)u|Transcribe a pending voice event and echo transcripts once → ``(enriched_text, transcripts)`` for ``agent.interrupt()`` or the pending-drain flow; ``(text, [])`` when there is no STT-eligible media (caller owns the ``_build_media_placeholder`` fallback). ``text`` is only that fallback: the caption is read live from ``event.text`` so one merged in during the STT await is kept.NrZz%s transcription failed: %s) rFrPr_thread_metadata_for_source_reply_anchor_for_eventrcr1r-warning) r3rDr`rarLrSrRrOrV trans_excs &&&&&$$ r"_transcribe_and_echo_pending_voiceCGatewayInboundTranscriptionMixin._transcribe_and_echo_pending_voice|s..u558O /3/X/XY^/_)_ &M6!;;FD`D`afDgh99h:  !((D+5 5 *`   NN8+ Q8O s]CBBAB;B< B B CBB CB<6C7C<CCz.(The user sent a message with no text content)c$V^8dQhRRRRRR/#)rprefixr r rr)rs"rrrs!333rc bV'd'VP4VP8wdV RV 2#V#)z\``prefix`` + the user's text; the Discord empty-content placeholder is dropped as redundant.r)strip_EMPTY_TEXT_PLACEHOLDER)clsrnr s&&&rr)6GatewayInboundTranscriptionMixin._prepend_media_prefixs1 *c.I.IIXT)- - rc V^8dQhRRRR/#)rr&r rr)rs"rrrsrrrrrc d^RIHpV!\PP V44pRV R2#)uOne minimal neutral marker for every STT failure. Never mention "no STT provider" or setup steps — persisted in history they make the model keep volunteering STT-setup advice.)to_agent_visible_cache_pathzR[voice message could not be transcribed automatically; the audio is available at: r)tools.credential_filesrvr%r&r')r&rv agent_paths& rr2:GatewayInboundTranscriptionMixin._untranscribed_audio_notes1 G01FG cdncoopqqrc V^8dQhRRRR/#)rr&r rztuple[Optional[str], str]r)rs"rrrs--s-kD-rc "\P!W!RR4GRjxL pVPR4'gN\P!W14GRjxL pVPR4'd\P RV4TpVR,'g:\P RWPRR44RVP V43#VR,pT;'gR P 4'gR #VR V R 23#LL5i) zS``(transcript_or_None, note)`` for one clip via configured STT with local fallback.Ngatewaysuccessz6Configured STT failed for %s; recovered with local STTz%Voice transcription failed for %s: %sr.z unknown errorr;r")Nu[The user sent a voice message but it came through empty or inaudible — speech-to-text returned no words. Do not guess at the content; ask the user to resend or type it out.])asyncio to_threadgetr-infor2rp)r3r&rrresultfallbackr;s&&&& rr05GatewayInboundTranscriptionMixin._transcribe_one_clips(()9yQQzz)$$$../NUUH||I&& TVZ[!i  KK?zzRY[jGk l77== =L)   b'')) Qzl!,,,-RUs?DDDDDD/&DA D!D8 DDr)N)__name__ __module__ __qualname____firstlineno____doc__r=rFrPrcrrkrq classmethodr) staticmethodr2r0__static_attributes__rrrrrsrM01d 5> JN ' OU,O rr--rr) r __future__rrloggingr%typingrgateway.run_commonr getLoggerr-rrrrrs8^" %   = )k-k-r