js-tts-wrapper
Version:
A JavaScript/TypeScript library that provides a unified API for working with multiple cloud-based Text-to-Speech (TTS) services
506 lines (498 loc) • 19 kB
JavaScript
;
var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
if (k2 === undefined) k2 = k;
var desc = Object.getOwnPropertyDescriptor(m, k);
if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
desc = { enumerable: true, get: function() { return m[k]; } };
}
Object.defineProperty(o, k2, desc);
}) : (function(o, m, k, k2) {
if (k2 === undefined) k2 = k;
o[k2] = m[k];
}));
var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
Object.defineProperty(o, "default", { enumerable: true, value: v });
}) : function(o, v) {
o["default"] = v;
});
var __importStar = (this && this.__importStar) || (function () {
var ownKeys = function(o) {
ownKeys = Object.getOwnPropertyNames || function (o) {
var ar = [];
for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
return ar;
};
return ownKeys(o);
};
return function (mod) {
if (mod && mod.__esModule) return mod;
var result = {};
if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
__setModuleDefault(result, mod);
return result;
};
})();
Object.defineProperty(exports, "__esModule", { value: true });
exports.SAPITTSClient = void 0;
const abstract_tts_1 = require("../core/abstract-tts");
const SpeechMarkdown = __importStar(require("../markdown/converter"));
const node_child_process_1 = require("node:child_process");
const node_fs_1 = require("node:fs");
const node_os_1 = require("node:os");
const node_path_1 = require("node:path");
/**
* SAPI TTS Client for Windows environments
*
* This client uses Windows Speech API (SAPI) through PowerShell to provide:
* - High-quality Windows TTS synthesis
* - SSML support
* - Rich voice metadata
* - Word boundary events
* - Rate, pitch, and volume controls
*/
class SAPITTSClient extends abstract_tts_1.AbstractTTSClient {
constructor(credentials = {}) {
super(credentials);
// Validate Windows environment
this.validateEnvironment();
// Set a default voice (will be determined at runtime)
this.voiceId = null;
}
/**
* Validate that we're running on Windows with PowerShell available
*/
validateEnvironment() {
// Check if we're in a Node.js environment
if (typeof process === "undefined" || !process.versions || !process.versions.node) {
throw new Error("SAPITTSClient is only supported in Node.js environments");
}
// Check if we're on Windows
if (process.platform !== "win32") {
throw new Error(`SAPITTSClient is only supported on Windows. Current platform: ${process.platform}`);
}
}
/**
* Get the list of required credential types for this engine
* @returns Array of required credential field names
*/
getRequiredCredentials() {
return []; // SAPI doesn't require credentials, only environment validation
}
/**
* SAPI does not require credentials but we validate the environment
*/
async checkCredentials() {
try {
this.validateEnvironment();
// Test if PowerShell and System.Speech are available
const testScript = `
try {
Add-Type -AssemblyName System.Speech
$synth = [System.Speech.Synthesis.SpeechSynthesizer]::new()
$voices = $synth.GetInstalledVoices()
Write-Output "OK"
} catch {
Write-Error $_.Exception.Message
exit 1
}
`;
const result = await this.runPowerShellScript(testScript);
return result.trim() === "OK";
}
catch (error) {
console.error("SAPI TTS not available:", error);
return false;
}
}
/**
* Get available SAPI voices with rich metadata
*/
async getVoices() {
try {
const script = `
Add-Type -AssemblyName System.Speech
$synth = [System.Speech.Synthesis.SpeechSynthesizer]::new()
$voices = $synth.GetInstalledVoices()
$voiceData = @()
foreach ($voice in $voices) {
if ($voice.Enabled) {
$info = $voice.VoiceInfo
$voiceObj = @{
Name = $info.Name
Id = $info.Id
Gender = $info.Gender.ToString()
Age = $info.Age.ToString()
Culture = $info.Culture.Name
Language = $info.Culture.DisplayName
Description = $info.Description
}
$voiceData += $voiceObj
}
}
$voiceData | ConvertTo-Json -Depth 3
`;
const result = await this.runPowerShellScript(script);
const parsedResult = JSON.parse(result);
// Ensure we have an array (PowerShell returns single object when there's only one voice)
const voiceData = Array.isArray(parsedResult) ? parsedResult : [parsedResult];
// Convert to unified voice format
const unifiedVoices = voiceData.map((voice) => ({
id: voice.Id || voice.Name || "unknown",
name: voice.Name || "Unknown Voice",
gender: voice.Gender === "Male" || voice.Gender === "Female" ? voice.Gender : "Unknown",
provider: "sapi",
languageCodes: [
{
bcp47: voice.Culture || "en-US",
iso639_3: this.convertCultureToISO639(voice.Culture || "en-US"),
display: voice.Language || "English (United States)",
},
],
}));
return unifiedVoices;
}
catch (error) {
console.error("Error getting SAPI voices:", error);
return [];
}
}
/**
* Get raw voices from SAPI
*/
async _getVoices() {
return this.getVoices();
}
/**
* Run a PowerShell script and return the output
*/
async runPowerShellScript(script) {
return new Promise((resolve, reject) => {
const powershell = (0, node_child_process_1.spawn)("powershell.exe", [
"-NoProfile",
"-NonInteractive",
"-ExecutionPolicy",
"Bypass",
"-Command",
script,
]);
let stdout = "";
let stderr = "";
powershell.stdout.on("data", (data) => {
stdout += data.toString();
});
powershell.stderr.on("data", (data) => {
stderr += data.toString();
});
powershell.on("close", (code) => {
if (code === 0) {
resolve(stdout);
}
else {
reject(new Error(`PowerShell script failed with code ${code}: ${stderr}`));
}
});
powershell.on("error", (error) => {
reject(new Error(`Failed to start PowerShell: ${error.message}`));
});
});
}
/**
* Convert culture code to ISO 639-3 language code
*/
convertCultureToISO639(culture) {
const cultureMap = {
"en-US": "eng",
"en-GB": "eng",
"en-AU": "eng",
"en-CA": "eng",
"es-ES": "spa",
"es-MX": "spa",
"fr-FR": "fra",
"fr-CA": "fra",
"de-DE": "deu",
"it-IT": "ita",
"pt-BR": "por",
"pt-PT": "por",
"ru-RU": "rus",
"ja-JP": "jpn",
"ko-KR": "kor",
"zh-CN": "cmn",
"zh-TW": "cmn",
"ar-SA": "ara",
"hi-IN": "hin",
"th-TH": "tha",
"vi-VN": "vie",
"nl-NL": "nld",
"sv-SE": "swe",
"da-DK": "dan",
"no-NO": "nor",
"fi-FI": "fin",
"pl-PL": "pol",
"cs-CZ": "ces",
"hu-HU": "hun",
"tr-TR": "tur",
"he-IL": "heb",
};
const langCode = culture.split("-")[0];
return cultureMap[culture] || cultureMap[langCode] || "eng";
}
/**
* Synthesize text to audio bytes using SAPI
*/
async synthToBytes(text, options) {
try {
// Create a temporary filename for the audio export
const tempFilename = (0, node_path_1.join)((0, node_os_1.tmpdir)(), `${SAPITTSClient.TEMP_PREFIX}${Date.now()}.wav`);
// Prepare synthesis options
const voice = options?.voice || this.voiceId || null;
const rate = this.convertRate(options?.rate);
const volume = this.convertVolume(options?.volume);
// Prepare text for synthesis (handle Speech Markdown and SSML)
let processedText = text;
// Convert from Speech Markdown if requested
if (options?.useSpeechMarkdown && SpeechMarkdown.isSpeechMarkdown(processedText)) {
const ssmlText = await SpeechMarkdown.toSSML(processedText, "microsoft-azure");
processedText = ssmlText;
}
// Always ensure SSML format - wrap plain text in speak tags like other engines
if (this._isSSML(processedText)) {
processedText = this.ensureProperSSML(processedText);
}
else {
// Wrap plain text in SSML tags to enable consistent SSML processing
processedText = this.ensureProperSSML(`<speak>${processedText}</speak>`);
}
const escapedText = this.escapePowerShellString(processedText);
// Always use SSML processing since we now always have SSML format
const isSSMLProcessed = true;
// Build PowerShell script for synthesis
const script = `
Add-Type -AssemblyName System.Speech
[Console]::InputEncoding = [System.Text.Encoding]::UTF8
$synth = [System.Speech.Synthesis.SpeechSynthesizer]::new()
# Set voice if specified
${voice ? `
try {
$synth.SelectVoice("${this.escapePowerShellString(voice)}")
} catch {
# If voice selection fails, continue with default voice
Write-Warning "Could not select voice '${this.escapePowerShellString(voice)}', using default voice"
}` : ""}
# Set speech properties
$synth.Rate = ${rate}
$synth.Volume = ${volume}
# Set output to WAV file
$synth.SetOutputToWaveFile("${this.escapePowerShellString(tempFilename)}")
try {
# Synthesize speech using SSML (all text is now wrapped in SSML format)
$synth.SpeakSsml($text)
Write-Output "SUCCESS"
} catch {
Write-Error $_.Exception.Message
exit 1
} finally {
$synth.Dispose()
}
`;
// Execute PowerShell script
const result = await this.runPowerShellScript(`$text = "${escapedText}"; ${script}`);
if (!result.includes("SUCCESS")) {
throw new Error("SAPI synthesis failed");
}
// Read the generated WAV file
if (!(0, node_fs_1.existsSync)(tempFilename)) {
throw new Error("SAPI failed to generate audio file");
}
const audioBuffer = (0, node_fs_1.readFileSync)(tempFilename);
// Clean up the temporary file
try {
(0, node_fs_1.unlinkSync)(tempFilename);
}
catch (cleanupError) {
console.warn("Could not clean up temporary file:", cleanupError);
}
// Create estimated word timings (SAPI doesn't provide real-time events in this mode)
this._createEstimatedWordTimings(isSSMLProcessed ? this.stripSSML(processedText) : processedText);
return new Uint8Array(audioBuffer);
}
catch (error) {
console.error("Error synthesizing speech with SAPI:", error);
throw error;
}
}
/**
* Synthesize text to a byte stream with word boundaries
*/
async synthToBytestream(text, options) {
try {
// Get the audio bytes first
const audioBytes = await this.synthToBytes(text, options);
// For now, use estimated word boundaries
// TODO: Implement real-time word boundary events using SAPI events
let processedText = text;
// Convert from Speech Markdown if requested
if (options?.useSpeechMarkdown && SpeechMarkdown.isSpeechMarkdown(processedText)) {
const ssmlText = await SpeechMarkdown.toSSML(processedText, "microsoft-azure");
processedText = ssmlText;
}
// Always ensure SSML format - wrap plain text in speak tags like other engines
if (this._isSSML(processedText)) {
processedText = this.ensureProperSSML(processedText);
}
else {
// Wrap plain text in SSML tags to enable consistent SSML processing
processedText = this.ensureProperSSML(`<speak>${processedText}</speak>`);
}
const plainText = this._isSSML(processedText) ? this.stripSSML(processedText) : processedText;
const words = plainText.split(/\s+/).filter((word) => word.length > 0);
const estimatedDuration = 0.3; // Estimated duration per word in seconds
const wordBoundaries = [];
let currentTime = 0;
for (const word of words) {
if (word.trim()) {
wordBoundaries.push({
text: word,
offset: currentTime * 1000, // Convert to milliseconds
duration: estimatedDuration * 1000, // Convert to milliseconds
});
currentTime += estimatedDuration;
}
}
// Create a readable stream from the audio bytes
const audioStream = new ReadableStream({
start(controller) {
controller.enqueue(audioBytes);
controller.close();
},
});
return {
audioStream,
wordBoundaries,
};
}
catch (error) {
console.error("Error synthesizing speech to stream with SAPI:", error);
throw error;
}
}
/**
* Convert rate option to SAPI rate format
* @param rate Rate option (string or number)
* @returns Rate value for SAPI (-10 to 10)
*/
convertRate(rate) {
if (!rate)
return 0; // Default rate
if (typeof rate === "string") {
switch (rate.toLowerCase()) {
case "x-slow":
return -8;
case "slow":
return -4;
case "medium":
case "normal":
return 0;
case "fast":
return 4;
case "x-fast":
return 8;
default:
// Try to parse as number
const parsed = Number.parseFloat(rate);
return Number.isNaN(parsed) ? 0 : Math.max(-10, Math.min(10, parsed));
}
}
if (typeof rate === "number") {
// Clamp between -10 and 10
return Math.max(-10, Math.min(10, rate));
}
return 0;
}
/**
* Convert volume option to SAPI volume format
* @param volume Volume option (string or number)
* @returns Volume value for SAPI (0 to 100)
*/
convertVolume(volume) {
if (!volume)
return 100; // Default volume
if (typeof volume === "string") {
switch (volume.toLowerCase()) {
case "silent":
return 0;
case "x-soft":
return 20;
case "soft":
return 40;
case "medium":
case "normal":
return 60;
case "loud":
return 80;
case "x-loud":
return 100;
default:
// Try to parse as number
const parsed = Number.parseFloat(volume);
return Number.isNaN(parsed) ? 100 : Math.max(0, Math.min(100, parsed));
}
}
if (typeof volume === "number") {
// Clamp between 0 and 100
return Math.max(0, Math.min(100, volume));
}
return 100;
}
/**
* Escape a string for use in PowerShell
* @param str String to escape
* @returns Escaped string
*/
escapePowerShellString(str) {
return str
.replace(/\\/g, "\\\\")
.replace(/"/g, '""')
.replace(/`/g, "``")
.replace(/\$/g, "`$");
}
/**
* Strip SSML tags from text (fallback for plain text processing)
* @param text Text with SSML tags
* @returns Plain text without SSML tags
*/
stripSSML(text) {
// Simple SSML tag removal - SAPI handles SSML natively, but this is for word timing estimation
return text.replace(/<[^>]*>/g, "").trim();
}
/**
* Ensure SSML has proper format for SAPI
* @param text SSML text
* @returns Properly formatted SSML
*/
ensureProperSSML(text) {
// Trim whitespace for proper parsing
const trimmedText = text.trim();
// Check if the SSML already has version attribute
if (trimmedText.includes('version=')) {
return text; // Return original text to preserve formatting
}
// If it's a simple <speak> tag, add the version attribute
// Note: SAPI requires xml:lang="en" (not "en-US") for SSML to work properly
if (trimmedText.startsWith('<speak>')) {
return text.replace('<speak>', '<speak version="1.0" xml:lang="en">');
}
// If it doesn't start with <speak>, wrap it properly
// Note: SAPI requires xml:lang="en" (not "en-US") for SSML to work properly
if (!trimmedText.startsWith('<speak')) {
return `<speak version="1.0" xml:lang="en">${text}</speak>`;
}
return text;
}
}
exports.SAPITTSClient = SAPITTSClient;
Object.defineProperty(SAPITTSClient, "TEMP_PREFIX", {
enumerable: true,
configurable: true,
writable: true,
value: "sapi_tts_"
});