Configuration Reference
Configuration schema for Persyk application settings
Type: object
Section titled “Type: object”⚠️ Additional properties are not allowed.
| Property | Type | Required | Possible values | Deprecated | Default | Description | Examples |
|---|---|---|---|---|---|---|---|
| providers | object | ✅ | object | {} | API provider configurations | ||
| selectedAudioDevice | string | string | ID of the selected audio input device | ||||
| transcription | object | object | Transcription settings | ||||
| transcription.enabled | boolean | ✅ | boolean | true | Whether transcription is enabled | ||
| transcription.provider | string | ✅ | string | Name of the provider to use for transcription | |||
| transcription.model | string | ✅ | string | "whisper-1" | Transcription model name | ||
| transcription.requestOptions | object | object | Additional transcription request options | ||||
| transcription.requestOptions.language | string | string | Language code for transcription (e.g., ‘en’, ‘es’) | ||||
| transcription.requestOptions.prompt | string | string | Optional prompt to guide the transcription style | ||||
| transcription.requestOptions.temperature | number | number | Sampling temperature between 0 and 1 | ||||
| transcription.copyToClipboard | boolean | ✅ | boolean | true | Copy transcription to clipboard after completion | ||
| transcription.notifications | boolean | ✅ | boolean | true | Show notifications for transcription events | ||
| transcription.formatting | object | object | LLM formatting configuration | ||||
| transcription.formatting.enabled | boolean | ✅ | boolean | true | Whether to format transcriptions using an LLM | ||
| transcription.formatting.provider | string | ✅ | string | Name of the provider to use for formatting | |||
| transcription.formatting.model | string | ✅ | string | Model to use for formatting | |||
| transcription.formatting.prompt | string | ✅ | string | "Format this voice memo transcription into clear, readable text. Apply the following improvements:\n\n- Structure: Organize the content into logical paragraphs\n- Grammar: Fix run-on sentences, add appropriate punctuation, and correct obvious grammatical errors\n- Filler words: Remove filler words like \"um,\" \"uh,\" \"like,\" \"you know\" (unless they seem intentional)\n- Clarity: Preserve the original meaning and tone while making it more readable\n- Lists: Convert rambling lists into bullet points where appropriate\n- Timestamps/sections: If there are distinct topic changes, consider adding section headers\n\nDo NOT:\n\n- Change the core meaning or message\n- Add information that wasn't in the original\n- Make it overly formal if the tone was casual\n- Remove personal voice characteristics that add authenticity\n- Include any preamble, explanation, or commentary\n- Add phrases like \"Here's the formatted version\" or \"I've cleaned up the text\"\n\nOUTPUT INSTRUCTIONS:\nReturn ONLY the formatted transcription itself. No introductory text, no explanations, no meta-commentary. Just the cleaned-up transcription.\n\nHere's the transcription to format:\n{{TRANSCRIPTION}}" | Prompt template for formatting (use {{TRANSCRIPTION}} as placeholder) | ||
| webhooks | object | object | Webhook configurations | ||||
| webhooks.transcription-complete | object | object | Webhook triggered when transcription completes | ||||
| webhooks.transcription-complete.enabled | boolean | ✅ | boolean | true | Whether this webhook is enabled | ||
| webhooks.transcription-complete.url | string | ✅ | Format: uri | Webhook endpoint URL | |||
| webhooks.transcription-complete.headers | object | object | Custom headers to include in the webhook request |
Markdown generated with jsonschema-markdown.