[
https://issues.apache.org/jira/browse/CAMEL-23390?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Work on CAMEL-23390 started by Federico Mariani.
------------------------------------------------
> Add audio transcription support to camel-openai component
> ---------------------------------------------------------
>
> Key: CAMEL-23390
> URL: https://issues.apache.org/jira/browse/CAMEL-23390
> Project: Camel
> Issue Type: New Feature
> Components: camel-openai
> Affects Versions: 4.20.0
> Reporter: Federico Mariani
> Assignee: Federico Mariani
> Priority: Major
>
> The camel-openai component currently supports three operations:
> _chat-completion_, _embeddings_, and _tool-execution_. The underlying
> openai-java SDK already provides full support for the _POST
> /v1/audio/transcriptions_ endpoint via
> _client.audio().transcriptions().create()_, including response models
> (Transcription, TranscriptionVerbose, TranscriptionDiarized) and streaming
> events.
>
>
>
>
> This issue proposes adding a new audio-transcription operation to
> camel-openai that:
>
>
>
> - Accepts an audio file as the exchange body (e.g., File, InputStream, or
> byte[])
>
>
> - Exposes key parameters as endpoint options and/or headers: language,
> prompt, responseFormat, temperature, timestampGranularities
>
>
> - Returns the transcription text as the exchange body, with metadata (e.g.,
> duration, segments) available via headers
>
>
>
>
>
> This would allow Camel routes to transcribe audio files using the OpenAI API
> without requiring custom code or a separate HTTP call.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
