mirror of
https://github.com/bilawalsidhu/gods-eye-view.git
synced 2026-11-04 10:24:02 +00:00
Manjunath
ecaddd838b
Both stack parents (#837, #842) are on main, so the voice geometry branch now builds on main directly. Conflict resolutions keep main's behaviour and add the geometry work on top: osm_query gating on every provider, restored admin identity for Nominatim outlines, merged analyst follow-up memory, and the cancelled Globe View reset instruction ahead of the generic cancellation.
58 lines
2.0 KiB
JavaScript
58 lines
2.0 KiB
JavaScript
import { realtimeInstructions } from '../openai/instructions.js';
|
|
import { GEV_REALTIME_TOOLS, offeredTools } from '../openai/tools.js';
|
|
|
|
export const GEMINI_LIVE_MODEL_DEFAULT = 'gemini-3.8-live';
|
|
export const GEMINI_TOKEN_ENDPOINT =
|
|
'https://generativelanguage.googleapis.com/v1beta/auth_tokens';
|
|
export const GEMINI_LIVE_WEBSOCKET_URL =
|
|
'wss://generativelanguage.googleapis.com/ws/google.ai.generativelanguage.v1beta.GenerativeService.BidiGenerateContentConstrained';
|
|
|
|
export function normalizeGeminiModel(value) {
|
|
const model = String(value ?? GEMINI_LIVE_MODEL_DEFAULT)
|
|
.trim()
|
|
.replace(/^models\//, '');
|
|
return /^gemini-[a-z0-9][a-z0-9.-]{0,119}$/.test(model) ? model : null;
|
|
}
|
|
|
|
/**
|
|
* REST BidiGenerateContentSetup fields, excluding its model resource name.
|
|
* `overpass` says an operator configured Overpass, which `osm_query` needs.
|
|
*/
|
|
export function createGeminiLiveConfig(
|
|
annotationGuidance,
|
|
inputMode = 'open-mic',
|
|
tools = GEV_REALTIME_TOOLS,
|
|
{ overpass = false } = {},
|
|
) {
|
|
if (!['open-mic', 'push-to-talk'].includes(inputMode))
|
|
throw new Error('Invalid Gemini input mode');
|
|
const config = {
|
|
generationConfig: { responseModalities: ['AUDIO'] },
|
|
systemInstruction: {
|
|
parts: [{ text: realtimeInstructions(annotationGuidance, { overpass }) }],
|
|
},
|
|
tools: [
|
|
{
|
|
functionDeclarations: offeredTools(tools, { overpass }).map(
|
|
({ name, description, parameters }) => ({
|
|
name,
|
|
description,
|
|
// GEV's canonical schemas are JSON Schema, not Google's Schema.
|
|
parametersJsonSchema: structuredClone(parameters),
|
|
// App actions must finish before the model confirms their result.
|
|
behavior: 'BLOCKING',
|
|
}),
|
|
),
|
|
},
|
|
],
|
|
inputAudioTranscription: {},
|
|
outputAudioTranscription: {},
|
|
contextWindowCompression: { slidingWindow: {} },
|
|
};
|
|
if (inputMode === 'push-to-talk')
|
|
config.realtimeInputConfig = {
|
|
automaticActivityDetection: { disabled: true },
|
|
};
|
|
return config;
|
|
}
|