feat: add STT support, Gemini TTS, and expand usage tracking
- Speech-to-Text: full pipeline with sttCore handler, /v1/audio/transcriptions endpoint, sttConfig for OpenAI, Gemini, Groq, Deepgram, AssemblyAI, HuggingFace, NVIDIA Parakeet; new 9router-stt skill - Gemini TTS: add gemini provider with 30 prebuilt voices and TTS_PROVIDER_CONFIG - Usage: implement GLM (intl/cn) and MiniMax (intl/cn) quota fetchers; refactor Gemini CLI usage to use retrieveUserQuota with per-model buckets - Disabled models: lowdb-backed disabledModelsDb + /api/models/disabled route - Header search: reusable Zustand store (headerSearchStore) wired into Header - CLI tools: add Claude Cowork tool card and cowork-settings API - Providers: introduce mediaPriority sorting in getProvidersByKind, add Kimi K2.6, reorder hermes, drop qwen STT kind - UI: expand media-providers/[kind]/[id] page (+314), enhance OAuthModal, ModelSelectModal, ProviderTopology, ProxyPools, ProviderLimits - Assets: refresh provider PNGs (alicode, byteplus, cloudflare-ai, nvidia, ollama, vertex, volcengine-ark) and add aws-polly, fal-ai, jina-ai, recraft, runwayml, stability-ai, topaz, black-forest-labs
This commit is contained in:
decolua
117
open-sse/handlers/ttsProviders/gemini.js
Normal file
117
open-sse/handlers/ttsProviders/gemini.js
Normal file
@@ -0,0 +1,117 @@
|
||||
// Gemini TTS — generateContent with AUDIO modality returns PCM L16, wrap as WAV
|
||||
import { Buffer } from "node:buffer";
|
||||
|
||||
const DEFAULT_MODEL = "gemini-2.5-flash-preview-tts";
|
||||
const DEFAULT_VOICE = "Kore";
|
||||
const KNOWN_MODELS = ["gemini-2.5-flash-preview-tts", "gemini-2.5-pro-preview-tts"];
|
||||
|
||||
// Parse "model/voice" — if input doesn't match a known TTS model, treat it as voice with default model
|
||||
function parseGeminiModelVoice(input) {
|
||||
if (!input) return { modelId: DEFAULT_MODEL, voiceId: DEFAULT_VOICE };
|
||||
for (const id of KNOWN_MODELS) {
|
||||
if (input === id) return { modelId: id, voiceId: DEFAULT_VOICE };
|
||||
if (input.startsWith(`${id}/`)) return { modelId: id, voiceId: input.slice(id.length + 1) };
|
||||
}
|
||||
return { modelId: DEFAULT_MODEL, voiceId: input };
|
||||
}
|
||||
// Gemini returns PCM 16-bit signed mono @ 24kHz
|
||||
const SAMPLE_RATE = 24000;
|
||||
const CHANNELS = 1;
|
||||
const BITS_PER_SAMPLE = 16;
|
||||
|
||||
// Build WAV header for raw PCM payload
|
||||
function pcmToWav(pcmBuffer) {
|
||||
const dataSize = pcmBuffer.length;
|
||||
const byteRate = SAMPLE_RATE * CHANNELS * BITS_PER_SAMPLE / 8;
|
||||
const blockAlign = CHANNELS * BITS_PER_SAMPLE / 8;
|
||||
const header = Buffer.alloc(44);
|
||||
header.write("RIFF", 0);
|
||||
header.writeUInt32LE(36 + dataSize, 4);
|
||||
header.write("WAVE", 8);
|
||||
header.write("fmt ", 12);
|
||||
header.writeUInt32LE(16, 16);
|
||||
header.writeUInt16LE(1, 20);
|
||||
header.writeUInt16LE(CHANNELS, 22);
|
||||
header.writeUInt32LE(SAMPLE_RATE, 24);
|
||||
header.writeUInt32LE(byteRate, 28);
|
||||
header.writeUInt16LE(blockAlign, 32);
|
||||
header.writeUInt16LE(BITS_PER_SAMPLE, 34);
|
||||
header.write("data", 36);
|
||||
header.writeUInt32LE(dataSize, 40);
|
||||
return Buffer.concat([header, pcmBuffer]);
|
||||
}
|
||||
|
||||
// Build TTS prompt: add "Say [in {language}]:" prefix to force TTS mode
|
||||
function buildPrompt(text, language) {
|
||||
if (/:\s/.test(text)) return text; // user already provided style instruction
|
||||
return language ? `Say in ${language}: ${text}` : `Say: ${text}`;
|
||||
}
|
||||
|
||||
export default {
|
||||
async synthesize(text, model, credentials, _responseFormat, opts = {}) {
|
||||
if (!credentials?.apiKey) throw new Error("No Gemini API key configured");
|
||||
const { modelId, voiceId } = parseGeminiModelVoice(model);
|
||||
const url = `https://generativelanguage.googleapis.com/v1beta/models/${modelId}:generateContent?key=${credentials.apiKey}`;
|
||||
const res = await fetch(url, {
|
||||
method: "POST",
|
||||
headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({
|
||||
contents: [{ parts: [{ text: buildPrompt(text, opts.language) }] }],
|
||||
generationConfig: {
|
||||
responseModalities: ["AUDIO"],
|
||||
speechConfig: { voiceConfig: { prebuiltVoiceConfig: { voiceName: voiceId } } },
|
||||
},
|
||||
}),
|
||||
});
|
||||
if (!res.ok) {
|
||||
const err = await res.json().catch(() => ({}));
|
||||
throw new Error(err?.error?.message || `Gemini TTS failed: ${res.status}`);
|
||||
}
|
||||
const data = await res.json();
|
||||
const b64 = data?.candidates?.[0]?.content?.parts?.find((p) => p.inlineData?.data)?.inlineData?.data;
|
||||
if (!b64) {
|
||||
const reason = data?.candidates?.[0]?.finishReason || data?.promptFeedback?.blockReason || "unknown";
|
||||
throw new Error(`Gemini TTS returned no audio (finishReason: ${reason}, voice: ${voiceId}, model: ${modelId})`);
|
||||
}
|
||||
const wav = pcmToWav(Buffer.from(b64, "base64"));
|
||||
return { base64: wav.toString("base64"), format: "wav" };
|
||||
},
|
||||
};
|
||||
|
||||
// Voice fetcher — return prebuilt voices (Gemini has no list API)
|
||||
const PREBUILT_VOICES = [
|
||||
{ id: "Zephyr", lang: "en", gender: "Female" },
|
||||
{ id: "Puck", lang: "en", gender: "Male" },
|
||||
{ id: "Charon", lang: "en", gender: "Male" },
|
||||
{ id: "Kore", lang: "en", gender: "Female" },
|
||||
{ id: "Fenrir", lang: "en", gender: "Male" },
|
||||
{ id: "Leda", lang: "en", gender: "Female" },
|
||||
{ id: "Orus", lang: "en", gender: "Male" },
|
||||
{ id: "Aoede", lang: "en", gender: "Female" },
|
||||
{ id: "Callirrhoe", lang: "en", gender: "Female" },
|
||||
{ id: "Autonoe", lang: "en", gender: "Female" },
|
||||
{ id: "Enceladus", lang: "en", gender: "Male" },
|
||||
{ id: "Iapetus", lang: "en", gender: "Male" },
|
||||
{ id: "Umbriel", lang: "en", gender: "Male" },
|
||||
{ id: "Algieba", lang: "en", gender: "Male" },
|
||||
{ id: "Despina", lang: "en", gender: "Female" },
|
||||
{ id: "Erinome", lang: "en", gender: "Female" },
|
||||
{ id: "Algenib", lang: "en", gender: "Male" },
|
||||
{ id: "Rasalgethi", lang: "en", gender: "Male" },
|
||||
{ id: "Laomedeia", lang: "en", gender: "Female" },
|
||||
{ id: "Achernar", lang: "en", gender: "Female" },
|
||||
{ id: "Alnilam", lang: "en", gender: "Male" },
|
||||
{ id: "Schedar", lang: "en", gender: "Male" },
|
||||
{ id: "Gacrux", lang: "en", gender: "Female" },
|
||||
{ id: "Pulcherrima", lang: "en", gender: "Female" },
|
||||
{ id: "Achird", lang: "en", gender: "Male" },
|
||||
{ id: "Zubenelgenubi", lang: "en", gender: "Male" },
|
||||
{ id: "Vindemiatrix", lang: "en", gender: "Female" },
|
||||
{ id: "Sadachbia", lang: "en", gender: "Male" },
|
||||
{ id: "Sadaltager", lang: "en", gender: "Male" },
|
||||
{ id: "Sulafat", lang: "en", gender: "Female" },
|
||||
];
|
||||
|
||||
export async function fetchGeminiVoices() {
|
||||
return PREBUILT_VOICES.map((v) => ({ voice_id: v.id, name: v.id, labels: { language: v.lang, gender: v.gender } }));
|
||||
}
|
||||
@@ -5,6 +5,7 @@ import localDevice, { fetchLocalDeviceVoices } from "./localDevice.js";
|
||||
import elevenlabs, { fetchElevenLabsVoices } from "./elevenlabs.js";
|
||||
import openai from "./openai.js";
|
||||
import openrouter from "./openrouter.js";
|
||||
import gemini, { fetchGeminiVoices } from "./gemini.js";
|
||||
import { FORMAT_HANDLERS } from "./genericFormats.js";
|
||||
import { parseModelVoice } from "./_base.js";
|
||||
|
||||
@@ -16,6 +17,7 @@ const SPECIAL_ADAPTERS = {
|
||||
elevenlabs,
|
||||
openai,
|
||||
openrouter,
|
||||
gemini,
|
||||
};
|
||||
|
||||
export function getTtsAdapter(provider) {
|
||||
@@ -41,7 +43,8 @@ export const VOICE_FETCHERS = {
|
||||
"edge-tts": fetchEdgeTtsVoices,
|
||||
"local-device": fetchLocalDeviceVoices,
|
||||
elevenlabs: fetchElevenLabsVoices,
|
||||
gemini: fetchGeminiVoices,
|
||||
};
|
||||
|
||||
// Re-export for backward compat
|
||||
export { fetchEdgeTtsVoices, fetchLocalDeviceVoices, fetchElevenLabsVoices };
|
||||
export { fetchEdgeTtsVoices, fetchLocalDeviceVoices, fetchElevenLabsVoices, fetchGeminiVoices };
|
||||
|
||||
Reference in New Issue
Block a user