echogarden 0.12.2 → 1.0.0

package/docs/Technical.md

@@ -16,7 +16,7 @@ Packages are downloaded as `.tar.gz` files, and are extracted to `[data-folder]/
 * `Users/User/Library/Application Support/echogarden` on macOS
 * `/home/user/.local/share/echogarden` on Linux
 
-By using downloadable packages, the installed size is made significantly smaller and the installation faster. The total size of all available packages is currently about 16.2GB (11.4GB compressed).
+By using downloadable packages, the installed size is made significantly smaller and the installation faster. The total size of all available packages is currently about 50GB (may not be up-to-date).
 
 The packages are currently hosted and downloaded from a dedicated [Hugging Face repository](https://huggingface.co/echogarden/echogarden-packages).
 
@@ -24,14 +24,14 @@ The packages are currently hosted and downloaded from a dedicated [Hugging Face
 
 The base installed (uncompressed) size, including dependencies, is around 270MB. This is without any models or voices, which are downloaded as needed.
 
-Currently, the biggest contributors to the size are:
+Currently, the largest contributors to the size are:
 
 * `onnxruntime-node` (NAPI): 92MB
 * `kuromoji` (JavaScript) 40MB
 * `flite-wasi` (WASI): 20MB
 * `espeak-ng-emscripten` (WASM): 18MB
 
-`onnxruntime-node` is big because it bundles pre-compiled binaries for multiple platforms. `kuromoji` is large because of its dictionary files and some unessential test code it bundles. The other three packages include large WASM binaries.
+`onnxruntime-node` is large because it bundles pre-compiled binaries for multiple platforms. `kuromoji` is large because of its dictionary files and some unessential test code it bundles. The other three packages include large WASM binaries.
 
 So, yes, in the future it may be possible to reduce the core installed size by dynamically installing some of these dependencies, or using modified, "slimmed-down" versions of some packages.
 
@@ -51,7 +51,7 @@ However, it is a lot of work, and only a subset of the engines can be supported
 * Possibly lots of issues with inconsistent browser support and browser security constraints
 * Not future-proof. Due to changing restrictions of browsers, the runtime environment is not guaranteed be reliably reproducible in the future, meaning that it may need continuous maintenance to ensure it keeps working on the newest browsers
 
-It remains to be seen if this sort of work would feel justified somehow. I designed the tool to make the local installation extremely easy and issue-free. I guess it could look "impressive" to be able to run it in a browser, and may be come as a nice "toy" or "tech-demo", and could get some attention, but it may eventually turn out to be simpler and more practical to just install a local instance and connect to it from the browser via a WebSocket API.
+It remains to be seen if this sort of work would feel justified somehow. I designed the tool to make the local installation extremely easy and issue-free. I guess it could look "impressive" to be able to run it in a browser, and may become as a nice "toy" or "tech-demo", and could get some attention, but it may eventually turn out to be simpler and more practical to just install a local instance and connect to it from the browser via a WebSocket API.
 
 A TTS-only browser extension is in development. It registers Echogarden's voices on the browser's Web Speech API using the [`chrome.ttsEngine`](https://developer.chrome.com/docs/extensions/reference/ttsEngine/) extension API and communicates with it using the WebSocket API
 

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "echogarden",
-	"version": "0.12.2",
+	"version": "1.0.0",
 	"description": "An integrated speech toolset, providing a variety of synthesis, recognition, alignment, and other processing tools, designed to be directly accessible to end-users.",
 	"author": "Rotem Dan",
 	"license": "GPL-3.0",
@@ -24,7 +24,7 @@
 		"url": "https://github.com/echogarden-project/echogarden/issues"
 	},
 	"engines": {
-		"node": ">=16"
+		"node": ">=18"
 	},
 	"os": [
 		"win32",
@@ -54,8 +54,8 @@
 		"echogarden": "./dist/cli/CLILauncher.js"
 	},
 	"dependencies": {
-		"@aws-sdk/client-polly": "^3.535.0",
-		"@aws-sdk/client-transcribe-streaming": "^3.535.0",
+		"@aws-sdk/client-polly": "^3.554.0",
+		"@aws-sdk/client-transcribe-streaming": "^3.554.0",
 		"@echogarden/espeak-ng-emscripten": "^0.1.2",
 		"@echogarden/fasttext-wasm": "^0.1.0",
 		"@echogarden/flite-wasi": "^0.1.1",
@@ -72,10 +72,10 @@
 		"chalk": "^5.3.0",
 		"cldr-segmentation": "^2.2.0",
 		"command-exists": "^1.2.9",
-		"compromise": "^14.12.0",
+		"compromise": "^14.13.0",
 		"compromise-dates": "^3.5.0",
 		"fs-extra": "^11.2.0",
-		"gaxios": "^6.3.0",
+		"gaxios": "^6.4.0",
 		"graceful-fs": "^4.2.11",
 		"html-escaper": "^3.0.3",
 		"html-to-text": "^9.0.5",
@@ -90,9 +90,11 @@
 		"ndarray": "^1.0.19",
 		"ndarray-ops": "^1.2.2",
 		"onnxruntime-node": "^1.17.0",
+		"openai": "^4.33.0",
 		"sam-js": "^0.2.1",
 		"strip-ansi": "^7.1.0",
-		"tar": "^6.2.0",
+		"tar": "^7.0.0",
+		"tiktoken": "^1.0.13",
 		"tinyld": "^1.3.4",
 		"ws": "^8.16.0",
 		"wtf_wikipedia": "^10.3.0"
@@ -121,14 +123,11 @@
 		"@types/msgpack-lite": "^0.1.11",
 		"@types/ndarray": "^1.0.14",
 		"@types/ndarray-ops": "^1.2.7",
-		"@types/node": "^20.11.28",
+		"@types/node": "^20.12.7",
 		"@types/recursive-readdir": "^2.2.4",
-		"@types/tar": "^6.1.11",
+		"@types/tar": "^6.1.12",
 		"@types/ws": "^8.5.10",
-		"@typescript-eslint/eslint-plugin": "^7.2.0",
-		"@typescript-eslint/parser": "^7.2.0",
-		"eslint": "^8.57.0",
-		"ts-json-schema-generator": "^1.5.0",
-		"typescript": "^5.4.2"
+		"ts-json-schema-generator": "^1.5.1",
+		"typescript": "^5.4.5"
 	}
 }

package/src/alignment/DTWMfccSequenceAlignment.ts

@@ -1,21 +1,21 @@
-import { cosineDistancePrecomputedMagnitudes, createVectorForIntegerRange, eucledianDistance, magnitude } from "../math/VectorMath.js"
-import { logToStderr } from "../utilities/Utilities.js"
-import { alignDTWWindowed } from "./DTWSequenceAlignmentWindowed.js"
+import { cosineDistancePrecomputedMagnitudes, createVectorForIntegerRange, euclidianDistance, magnitude } from '../math/VectorMath.js'
+import { logToStderr } from '../utilities/Utilities.js'
+import { alignDTWWindowed } from './DTWSequenceAlignmentWindowed.js'
 
 const log = logToStderr
 
-export async function alignMFCC_DTW(mfccFrames1: number[][], mfccFrames2: number[][], windowLength: number, distanceFunction: "eucledian" | "cosine" = "eucledian", centerIndexes?: number[]) {
-	if (distanceFunction == "eucledian") {
+export async function alignMFCC_DTW(mfccFrames1: number[][], mfccFrames2: number[][], windowLength: number, distanceFunction: 'euclidian' | 'cosine' = 'euclidian', centerIndexes?: number[]) {
+	if (distanceFunction == 'euclidian') {
 		const { path } = alignDTWWindowed(
 			mfccFrames1,
 			mfccFrames2,
-			eucledianDistance,
+			euclidianDistance,
 			windowLength,
 			centerIndexes
 		)
 
 		return path
-	} else if (distanceFunction == "cosine") {
+	} else if (distanceFunction == 'cosine') {
 		const indexes1 = createVectorForIntegerRange(0, mfccFrames1.length)
 		const indexes2 = createVectorForIntegerRange(0, mfccFrames2.length)
 
@@ -32,12 +32,12 @@ export async function alignMFCC_DTW(mfccFrames1: number[][], mfccFrames2: number
 
 		return path
 	} else {
-		throw new Error("Invalid distance function")
+		throw new Error('Invalid distance function')
 	}
 }
 
 export function getCostMatrixMemorySizeMB(sequence1Length: number, sequence2Length: number, windowLength: number) {
 	const costMatrixMemorySizeMB = sequence1Length * Math.min(sequence2Length, windowLength) * 4 / 1000000
-	
+
 	return costMatrixMemorySizeMB
 }

package/src/alignment/DTWSequenceAlignment.ts

@@ -1,5 +1,5 @@
-import { logToStderr } from "../utilities/Utilities.js"
-import { AlignmentPath } from "./SpeechAlignment.js"
+import { logToStderr } from '../utilities/Utilities.js'
+import { AlignmentPath } from './SpeechAlignment.js'
 
 const log = logToStderr
 

package/src/alignment/DTWSequenceAlignmentWindowed.ts

@@ -1,11 +1,11 @@
-import { logToStderr } from "../utilities/Utilities.js"
-import { AlignmentPath } from "./SpeechAlignment.js"
+import { logToStderr } from '../utilities/Utilities.js'
+import { AlignmentPath } from './SpeechAlignment.js'
 
 const log = logToStderr
 
 export function alignDTWWindowed<T, U>(sequence1: T[], sequence2: U[], costFunction: (a: T, b: U) => number, windowMaxLength: number, centerIndexes?: number[]) {
 	if (windowMaxLength < 2) {
-		throw new Error("Window length must be greater or equal to 2")
+		throw new Error('Window length must be greater or equal to 2')
 	}
 
 	if (sequence1.length == 0 || sequence2.length == 0) {

package/src/alignment/LevenshteinSequenceAlignment.ts

@@ -1,5 +1,5 @@
-import { logToStderr } from "../utilities/Utilities.js"
-import { AlignmentPath } from "./SpeechAlignment.js"
+import { logToStderr } from '../utilities/Utilities.js'
+import { AlignmentPath } from './SpeechAlignment.js'
 
 const log = logToStderr