@sign-speak/react-sdk
Version:
Unlock Sign Language Recognition, Avatar, and Speech Recognition.
297 lines (296 loc) • 14 kB
JavaScript
"use strict";
var __assign = (this && this.__assign) || function () {
__assign = Object.assign || function(t) {
for (var s, i = 1, n = arguments.length; i < n; i++) {
s = arguments[i];
for (var p in s) if (Object.prototype.hasOwnProperty.call(s, p))
t[p] = s[p];
}
return t;
};
return __assign.apply(this, arguments);
};
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
return new (P || (P = Promise))(function (resolve, reject) {
function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
step((generator = generator.apply(thisArg, _arguments || [])).next());
});
};
var __generator = (this && this.__generator) || function (thisArg, body) {
var _ = { label: 0, sent: function() { if (t[0] & 1) throw t[1]; return t[1]; }, trys: [], ops: [] }, f, y, t, g;
return g = { next: verb(0), "throw": verb(1), "return": verb(2) }, typeof Symbol === "function" && (g[Symbol.iterator] = function() { return this; }), g;
function verb(n) { return function (v) { return step([n, v]); }; }
function step(op) {
if (f) throw new TypeError("Generator is already executing.");
while (g && (g = 0, op[0] && (_ = 0)), _) try {
if (f = 1, y && (t = op[0] & 2 ? y["return"] : op[0] ? y["throw"] || ((t = y["return"]) && t.call(y), 0) : y.next) && !(t = t.call(y, op[1])).done) return t;
if (y = 0, t) op = [op[0] & 2, t.value];
switch (op[0]) {
case 0: case 1: t = op; break;
case 4: _.label++; return { value: op[1], done: false };
case 5: _.label++; y = op[1]; op = [0]; continue;
case 7: op = _.ops.pop(); _.trys.pop(); continue;
default:
if (!(t = _.trys, t = t.length > 0 && t[t.length - 1]) && (op[0] === 6 || op[0] === 2)) { _ = 0; continue; }
if (op[0] === 3 && (!t || (op[1] > t[0] && op[1] < t[3]))) { _.label = op[1]; break; }
if (op[0] === 6 && _.label < t[1]) { _.label = t[1]; t = op; break; }
if (t && _.label < t[2]) { _.label = t[2]; _.ops.push(op); break; }
if (t[2]) _.ops.pop();
_.trys.pop(); continue;
}
op = body.call(thisArg, _);
} catch (e) { op = [6, e]; y = 0; } finally { f = t = 0; }
if (op[0] & 5) throw op[1]; return { value: op[0] ? op[1] : void 0, done: true };
}
};
Object.defineProperty(exports, "__esModule", { value: true });
exports.runRequest = exports.genMedia = exports.Location = exports.produceSpeech = exports.produceSign = exports.submitFeedback = exports.recognizeSpeech = exports.recognizeSign = void 0;
var key_1 = require("./key");
var API_BASE = "https://api.sign-speak.com";
// Utility function for delaying polling iterations
function delay(ms) {
return new Promise(function (resolve) { return setTimeout(resolve, ms); });
}
/**
* Recognize sign language from a base64-encoded video.
* @param vidB64 Base64-encoded video string.
* @param options Optional parameters (e.g., model, language, additional customizations).
* @returns The full recognition result containing feedback and detailed predictions.
*/
function recognizeSign(vidB64, options) {
var _a;
return __awaiter(this, void 0, void 0, function () {
var payload, res;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
payload = __assign({ payload: vidB64, single_recognition_mode: true, request_class: "BLOCKING", model: (_a = options === null || options === void 0 ? void 0 : options.model) !== null && _a !== void 0 ? _a : "SLR.2.sm" }, options);
return [4 /*yield*/, runRequest("/recognize-sign", payload)];
case 1:
res = _b.sent();
return [2 /*return*/, res];
}
});
});
}
exports.recognizeSign = recognizeSign;
/**
* Recognize speech from a base64-encoded audio.
* @param audioB64 Base64-encoded audio string.
* @param options Optional parameters (e.g., model, language, additional customizations).
* @returns The full recognition result containing feedback and detailed predictions.
*/
function recognizeSpeech(audioB64, options) {
var _a, _b;
return __awaiter(this, void 0, void 0, function () {
var payload, res;
return __generator(this, function (_c) {
switch (_c.label) {
case 0:
payload = __assign({ payload: audioB64, single_recognition_mode: true, request_class: "BLOCKING", model: (_a = options === null || options === void 0 ? void 0 : options.model) !== null && _a !== void 0 ? _a : "LATEST", language: (_b = options === null || options === void 0 ? void 0 : options.language) !== null && _b !== void 0 ? _b : "en" }, options);
return [4 /*yield*/, runRequest("/recognize-speech", payload)];
case 1:
res = _c.sent();
return [2 /*return*/, res];
}
});
});
}
exports.recognizeSpeech = recognizeSpeech;
function submitFeedback(feedbackId, rating, correction) {
return __awaiter(this, void 0, void 0, function () {
var requestHeaders, payload, options;
return __generator(this, function (_a) {
switch (_a.label) {
case 0:
// If neither feedback nor correction is provided, there's nothing to submit.
if (!rating && !correction) {
return [2 /*return*/];
}
requestHeaders = new Headers();
requestHeaders.set("Content-Type", "application/json");
requestHeaders.set("X-api-key", (0, key_1.getKey)());
payload = {
rating: rating || "",
correction: correction || ""
};
options = {
method: "POST",
headers: requestHeaders,
body: JSON.stringify(payload)
};
return [4 /*yield*/, fetch("".concat(API_BASE, "/feedback/").concat(feedbackId), options)];
case 1:
_a.sent();
return [2 /*return*/];
}
});
});
}
exports.submitFeedback = submitFeedback;
/**
* Produce sign language video.
* Supports providing either a plain english string or timestamped english.
* @param input An object containing either `english` or `timestamped_english` (or both).
* @param options Optional parameters (e.g., model or additional customizations).
* @returns A Blob containing the generated sign video.
*/
function produceSign(input, options) {
return __awaiter(this, void 0, void 0, function () {
var payload;
return __generator(this, function (_a) {
switch (_a.label) {
case 0:
if (options == null) {
options = {};
}
if (options.model == undefined) {
options.model = "MALE";
}
payload = __assign(__assign({ request_class: "BLOCKING", resolution: 512 }, input), options);
return [4 /*yield*/, runRequest("/produce-sign", payload, "blob")];
case 1: return [2 /*return*/, _a.sent()];
}
});
});
}
exports.produceSign = produceSign;
/**
* Produce speech audio.
* Supports providing either a plain english string or timestamped english.
* @param input An object containing either `english` or `timestamped_english` (or both).
* @param options Optional parameters (e.g., model, additional customizations).
* @returns A Blob containing the generated speech audio.
*/
function produceSpeech(input, options) {
return __awaiter(this, void 0, void 0, function () {
var payload;
return __generator(this, function (_a) {
switch (_a.label) {
case 0:
if (options == null) {
options = {};
}
if (options.model == undefined) {
options.model = "MALE";
}
payload = __assign(__assign({ request_class: "BLOCKING" }, input), options);
return [4 /*yield*/, runRequest("/produce-speech", payload, "blob")];
case 1: return [2 /*return*/, _a.sent()];
}
});
});
}
exports.produceSpeech = produceSpeech;
/** Location enum for caption placement */
var Location;
(function (Location) {
Location["TOP"] = "TOP";
Location["BOTTOM"] = "BOTTOM";
})(Location = exports.Location || (exports.Location = {}));
/**
* Generate media (e.g., video) from a complex media generation request.
* Accepts a MediaGenRequest payload and returns a Blob containing the generated media.
* @param request A MediaGenRequest object.
* @param options Optional additional parameters for customization.
* @returns A Blob containing the generated media.
*/
function genMedia(request, options) {
return __awaiter(this, void 0, void 0, function () {
var payload;
return __generator(this, function (_a) {
switch (_a.label) {
case 0:
payload = __assign(__assign({}, request), options);
return [4 /*yield*/, runRequest("/gen-media", payload, "blob")];
case 1: return [2 /*return*/, _a.sent()];
}
});
});
}
exports.genMedia = genMedia;
/** ***************** Internal Request Handler ******************** */
/**
* Internal helper to run a request to the Sign-Speak API.
* This function also supports polling for asynchronous operations.
*
* @param endpoint REST endpoint (e.g., "/recognize-sign")
* @param payload Payload for the request.
* @param responseType Expected response type ("json" or "blob").
* @param pollInterval Polling interval in milliseconds.
* @param maxPolls Maximum number of polling attempts.
* @returns The API response.
*/
function runRequest(endpoint, payload, responseType, pollInterval, maxPolls) {
if (responseType === void 0) { responseType = "json"; }
if (pollInterval === void 0) { pollInterval = 1000; }
if (maxPolls === void 0) { maxPolls = 30; }
return __awaiter(this, void 0, void 0, function () {
var requestHeaders, requestOptions, response, data, batchId, polls;
return __generator(this, function (_a) {
switch (_a.label) {
case 0:
requestHeaders = new Headers();
requestHeaders.set("Content-Type", "application/json");
requestHeaders.set("X-api-key", (0, key_1.getKey)());
requestOptions = {
method: "POST",
headers: requestHeaders,
body: JSON.stringify(payload),
};
return [4 /*yield*/, fetch("".concat(API_BASE).concat(endpoint), requestOptions)];
case 1:
response = _a.sent();
if (!response.ok && response.status !== 202) {
throw new Error("Request failed with status ".concat(response.status));
}
if (!(responseType === "blob")) return [3 /*break*/, 3];
return [4 /*yield*/, response.blob()];
case 2:
data = _a.sent();
return [3 /*break*/, 5];
case 3: return [4 /*yield*/, response.json()];
case 4:
data = _a.sent();
_a.label = 5;
case 5:
if (!(response.status === 202)) return [3 /*break*/, 15];
batchId = data.batch_id;
polls = 0;
_a.label = 6;
case 6:
if (!(polls < maxPolls)) return [3 /*break*/, 14];
return [4 /*yield*/, delay(pollInterval)];
case 7:
_a.sent();
return [4 /*yield*/, fetch("".concat(API_BASE).concat(endpoint, "/").concat(batchId), { headers: requestHeaders })];
case 8:
response = _a.sent();
if (!(response.ok && response.status !== 202)) return [3 /*break*/, 13];
if (!(responseType === "blob")) return [3 /*break*/, 10];
return [4 /*yield*/, response.blob()];
case 9:
data = _a.sent();
if (data.size === 0) {
polls++;
return [3 /*break*/, 6];
}
return [3 /*break*/, 12];
case 10: return [4 /*yield*/, response.json()];
case 11:
data = _a.sent();
_a.label = 12;
case 12: return [2 /*return*/, data];
case 13:
polls++;
return [3 /*break*/, 6];
case 14: throw new Error("Polling exceeded maximum attempts");
case 15: return [2 /*return*/, data];
}
});
});
}
exports.runRequest = runRequest;