VidflowInstall

Summary

Fetch Kuaishou hot board videos up to a specified limit, download each video, extract its audio, transcribe the audio using BCut, and output the transcription along with optional video binary data.

Inputs

• Board Type — Number identifying the Kuaishou hot board type for automation; default is 1.
• Board ID — Number identifying the Kuaishou hot board ID for automation; default is 1.
• Max Videos — Maximum number of hot videos to automatically process in one run; accepts between 1 and 10, default is 3.
• Include Downloaded Video Binary — Boolean flag indicating whether to include the downloaded video files as binary outputs; default is false.
• FFmpeg Path — File path to the FFmpeg executable used for extracting audio from downloaded videos; defaults to '/usr/bin/ffmpeg'.
• Polling Interval Seconds — Number of seconds to wait between polling requests to BCut for transcription task status; defaults to 1 second.
• Timeout Seconds — Maximum number of seconds to wait before timing out the transcription task; defaults to 500 seconds.

Output shape

a single JSON object for each workflow input item containing data about the batch processing, including transcription results per video and optionally video binary data

The output JSON includes metadata for each processed video such as video ID, download URL, metrics (like counts, comments, plays), transcription text and segments, and status. If enabled, video binary data is included keyed by 'video_1', 'video_2', etc. Processing proceeds sequentially for up to the maximum specified videos from the selected hot board. Errors in individual video processing are captured in the output status without stopping the entire operation.

Links

• n8n community nodes installation guide