uPj)UdZddlmZmZddlmZmZddlmZm Z ddl m Z ddl m Z mZmZmZmZmZddlmZmZddlmZmZgd Zeeeed <gd Zeeeed <Gd de Zy)a Translates between OpenAI's `/v1/audio/transcriptions` shape and Soniox's async transcription API (https://soniox.com/docs/stt/async/async-transcription). This config covers parameter mapping, env validation and response shaping. The actual orchestration (file upload -> create -> poll -> fetch -> cleanup) lives in `litellm.llms.soniox.audio_transcription.handler`, because Soniox's async API requires multiple HTTP calls and does not fit the single-request contract of `base_llm_http_handler.audio_transcriptions`. )AnyFinal)HeadersResponse)AudioTranscriptionRequestDataBaseAudioTranscriptionConfig)BaseLLMException)SonioxExceptionget_soniox_api_baseget_soniox_api_keyrender_soniox_tokensrender_soniox_tokens_as_srtrender_soniox_tokens_as_vtt)AllMessageValues&OpenAIAudioTranscriptionOptionalParams) FileTypesTranscriptionResponse) language_hintslanguage_hints_strictenable_language_identificationenable_speaker_diarizationcontext translationclient_reference_id webhook_urlwebhook_auth_header_namewebhook_auth_header_value audio_urlfile_idSONIOX_PASSTHROUGH_PARAMS)soniox_polling_intervalsoniox_max_polling_attemptssoniox_cleanupfilenameSONIOX_HANDLER_ONLY_PARAMScHeZdZdZdedeefdZdededede def dZ d ed e d ee zde fd Z dd ededeedededed zded zdefdZ dded zded zdededede d zdefdZdededededef dZ ddeded zdefdZ ddeeefded zded zdefdZy )SonioxAudioTranscriptionConfigz$ ?r1 error_message status_codeheadersct|||S)Nmessager<r=)r )r.r;r<r=s r/get_error_classz.SonioxAudioTranscriptionConfig.get_error_classas}+W^__r1Nmessageslitellm_paramsapi_keyapi_basecpt|}|stddddd|i} |r| j|| S)NzsMissing Soniox API key. Set the SONIOX_API_KEY environment variable or pass api_key=... to litellm.transcription().ir? AuthorizationzBearer )r r update) r.r=r(rBr3rCrDrE resolved_keymerged_headerss r/validate_environmentz3SonioxAudioTranscriptionConfig.validate_environmentdsZ19 !O    w|n51    ! !' *r1streamct|SN)r )r.rErDr(r3rCrLs r/get_complete_urlz/SonioxAudioTranscriptionConfig.get_complete_urls#8,,r1 audio_filecnd|i}tD]}|j|}||||<t|ddS)a Build the JSON body for `POST /v1/transcriptions`. The handler is responsible for the file upload (if `audio_file` is bytes) and for filling in `file_id`/`audio_url`. This method exists so the config can be exercised in isolation by unit tests. r(Nzapplication/json)datafiles content_type)r r6r)r.r(rPr3rCbodyr9values r/%transform_audio_transcription_requestzDSonioxAudioTranscriptionConfig.transform_audio_transcription_requestsN(/&6, "C#'',E !S  " -$dQcddr1 raw_responsemodel_responsec  |j}|j ||S#t$r,}td|t|ddt|ddd}~wwxYw)az Build a TranscriptionResponse from a Soniox transcript payload. `raw_response.json()` may be either: - a Soniox transcript object: `{"id": "...", "text": "...", "tokens": [...]}` - or a merged envelope: `{"transcription": {...}, "transcript": {...}}` produced by the handler so transcription metadata is also available. z!Failed to parse Soniox response: r<ir=Nr?)rY)json Exceptionr getattr_build_response_from_payload)r.rXrYpayloadexcs r/&transform_audio_transcription_responsezESonioxAudioTranscriptionConfig.transform_audio_transcription_responsesn )..0G000XX !;C5A#L-E i>  s% A'AAr_r,ci}t|tr/d|vr+|jdxsi}|jdxsi}nt|tr|ni}|jdxsg}|dk(r|r t|}n}|dk(r|r t |}njt d|D}t d|D} |s| r|r t |}n4|jdr|d}n|rd jd |D}nd }|xs t| } || _ d | d <|jd t|ddz | d<t d|D} | rU|D chc]%} | jds| jd'} } t| dk(rtt| | d<|dk(r|rg} |D]q}d|jdd i}|jdt|ddz |d<|jdt|ddz |d<| j!|s| r| | d<| j"j%d||di| S#ttf$rY#wxYwcc} w)z:Shared response-building logic (also used by the handler). transcript transcriptiontokenssrtvttc3BK|]}|jdduyw)speakerNr6.0ts r/ zNSonioxAudioTranscriptionConfig._build_response_from_payload..s$RaQUU9%5T%A$Rc3BK|]}|jdduywr+Nrjrks r/rnzNSonioxAudioTranscriptionConfig._build_response_from_payload..sMquuZ0<Mrotextc3@K|]}|jddyw)rrrsNrjrks r/rnzNSonioxAudioTranscriptionConfig._build_response_from_payload..sAQquuVR0As)rr transcribetaskaudio_duration_msg@@durationc3BK|]}|jdduywrqrjrks r/rnzNSonioxAudioTranscriptionConfig._build_response_from_payload..sIQ155,D8Iror+ verbose_jsonwordstart_msstartend_msendwords soniox_raw)rdrc) isinstancedictr6rranyr joinrrrfloat TypeError ValueErrorlennextiterappend_hidden_paramsrH)r.r_rYr,transcription_metarcrerr has_speaker has_languageresponserm languagesrtoken word_entrys r/r^z;SonioxAudioTranscriptionConfig._build_response_from_payloads.0 gt $)@!(_!=!C  \28bJ$.w$=2J.8nnX.F.L" e #.v6D  %&.v6D"%$R6$R!RKMfMML|+F3'!&)wwA&AA(L,At,L '  ! !"5 6 B ',-?@S-T'UX^'^$ I&II ;AWaQUU:EVj 1WIW9~"'+DO'<$ n ,13E ).4eii6K-L 99Z(4*/j0A*BV*KJw'99X&2(-eHo(>(GJu% Z(  )$)! &&%7",  Cz*   Xs)II$,I$I! I!)NNrN)__name__ __module__ __qualname____doc__strr7rr0rboolr:intrr rArrKrOrrrWrrrarr^r-r1r/r'r'=sF//>d9e/       6`S`s`TT[^``p`##'(     t* F# -* -t - -  -  -t  -  -eee e  e ' e48<YY.4Y  Y68<&* Sc3hS.4St S  Sr1r'N)rtypingrrhttpxrr8litellm.llms.base_llm.audio_transcription.transformationrr)litellm.llms.base_llm.chat.transformationr litellm.llms.soniox.common_utilsr r r r rrlitellm.types.llms.openairrlitellm.types.utilsrrr r7r__annotations__r%r'r-r1r/rsq #GA /5c+  0E$s),R%ARr1