Communicates with OpenAI API (ChatGPT); the default MODEL is “gpt-4” (configurable). Supports almost all Chat features, such as re-asking based on histories and/or parameters; sampling temperature-Top_P, Presence-Frequency penalty, Logit bias.
Transcribes audio and video files in caption format WebVTT, using the OpenAI API “whisper-1” model by default (configurable) to convert audio data into text data. Set abbreviations and technical terms as PROMPT for more accurate transcription.
Translates audio into English. English text data is generated from sound or video file containing Japanese or other, using the OpenAI API “whisper-1” model. Abbreviations and technical terms can be set as PRMOPT to achieve more accurate translation.
Transcribes from audio file. The “whisper-1” model in the OpenAI API converts speech data into text data. If set abbreviations and technical terms as summary (PRMOPT), more accurate data conversion will be achieved.
Posts the text to OpenChat in the Workflow Platform. Since the ProcessID hashtag (topic) like “#p123” is automatically added, it will be displayed on the timeline of the process participants. Note that all users can browse with searching.
Posts to Open Chat (Enterprise Social in Questetra BPMS) with PID like “#p123” from the workflow engine standpoint. Posts will only be displayed on the participant timelines. Note that any BPMS user can search in their timeline, unlike Workflow Data.
Outputs business data to the Auto-Processing Log. Administrators of the Workflow App can verify the data that was stored when the Process reached this step.