PodfeedInstall

Summary

Generates podcast-style audio from text, a URL, a topic, a custom script, or uploaded files using the Podfeed AI service. The output can be a monologue (single voice) or a dialogue (two voices) in the selected language, with optional voice instructions and additional content controls.

Inputs

• inputType (required) — Source of content to convert to audio: Text, Website URL, Topic (research‑based), Script (pre‑written), or File (URLs/Base64). Determines which content‑specific field is shown.
• textContent (required) — The text to convert to audio. Shown only when inputType is Text.
• websiteUrl (required) — Website URL to scrape and convert to audio. Shown only when inputType is URL.
• topic (required) — Topic to research and generate content about. Shown only when inputType is Topic.
• scriptContent (required) — Pre‑written script to convert to audio. Shown only when inputType is Script.
• fileContent (required) — File URLs (https://…) or GCS URIs (gs://…), one per line, or Base64‑encoded file content. Shown only when inputType is File.
• mode (required) — Whether the audio should be a monologue (single voice) or a dialogue (two‑voice conversation).
• voice (required) — Voice ID for monologue mode. If omitted, defaults to 'gemini‑puck'. Options depend on the selected language.
• hostVoice (required) — Voice ID for the host in dialogue mode. If omitted, defaults to 'gemini‑puck'. Options depend on the selected language.
• cohostVoice (required) — Voice ID for the co‑host in dialogue mode. If omitted, defaults to 'gemini‑aoede'. Options depend on the selected language.
• language — Language of the generated content (e.g., en‑US, fr‑FR). Defaults to en‑US. Voice lists are filtered by this language.
• additionalOptions — Collection of optional settings that refine the audio generation.
• additionalOptions.userInstructions — Custom instructions for content generation. Hidden when inputType is Script.
• additionalOptions.cohostVoiceInstructions — Custom instructions for the co‑host voice (dialogue mode).
• additionalOptions.length — Desired length of the generated content: Short, Medium, or Long. Defaults to Medium. Hidden when inputType is Script.
• additionalOptions.level — Target audience level: Beginner, Intermediate, or Expert. Defaults to Intermediate. Hidden when inputType is Script.
• additionalOptions.emphasis — Content emphasis or focus area. Hidden when inputType is Script.
• additionalOptions.hostVoiceInstructions — Custom instructions for the host voice (dialogue mode).
• additionalOptions.questions — Include Q&A segments in the generated content. Hidden when inputType is Script.
• additionalOptions.readMode — When true, uses direct text reading mode (monologue only). Defaults to false.
• additionalOptions.voiceInstructions — Custom instructions for voice synthesis (monologue mode).

Output shape

object

The node returns the raw JSON response from the Podfeed POST /api/v1/audios endpoint. This includes metadata about the created audio generation task (e.g., taskId, status, and eventually the audio file URL). Subsequent nodes such as 'Get Task Status' or 'Get Audio' can use the returned IDs to poll for completion or retrieve the audio file.

Examples

Example 1: Generate a short English monologue from a text snippet

inputType=Text, textContent='Hello world', mode=Monologue, voice=en‑US‑standard‑A, language=en‑US, additionalOptions.length=Short

Example 2: Create a dialogue in Spanish from a website article

inputType=URL, websiteUrl='https://example.com/news', mode=Dialogue, hostVoice=es‑ES‑standard‑B, cohostVoice=es‑MX‑standard‑C, language=es‑ES