Wildbots GeminiInstall

Summary

Transcribe audio recordings to text using a selected Gemini AI model, accepting either an audio URL or a binary audio file as input.

Inputs

• Model (required) — Select the AI model to use for transcription, either from a searchable list or by ID.
• Input Type (required) — Choose whether the audio input is provided as a URL or as binary file data.
• URL (required) — The URL of the audio file to transcribe (required if Input Type is 'Audio URL').
• Input Data Field Name (required) — The name of the binary field containing audio data to process (required if Input Type is 'Binary File').
• Simplify Output — Optional boolean to return a simplified response format.
• Options — Optional settings including partial transcription start and end times (seconds) and maximum output token length.

Output shape

a list of transcription candidate objects or a single complex response object depending on 'Simplify Output' setting

If 'Simplify Output' is true, the node returns an array of simplified transcription candidates each mapped to the input item; otherwise, the full raw API response is returned in a single item array. Partial transcription times control the segment of audio transcribed. Errors are thrown on failure.

Examples

Example 1: Transcribing speech from a remote MP3 file

Select a Gemini audio model, set Input Type to 'Audio URL', provide the audio URL, adjust optional transcription start/end times if needed, and enable Simplify Output for easy usage.

Example 2: Transcribing speech from an uploaded audio file

Select a Gemini audio model, set Input Type to 'Binary File', specify the binary property name holding the audio data in the input item, and optionally configure partial transcription times and output simplification.

Links

• n8n community nodes installation guide