speechflow 1.3.2 → 1.4.0

package/CHANGELOG.md

@@ -2,6 +2,23 @@
 ChangeLog
 =========
 
+1.4.0 (2025-08-05)
+------------------
+
+- IMPROVEMENT: add dashboard infrastructure and expose information from "meter" and "trace" nodes
+- IMPROVEMENT: add subtitle web interface for real-time subtitle display
+- IMPROVEMENT: allow "deepgram" node to optionally provide "interim" results (primarily for subtitle)
+- IMPROVEMENT: allow "filter" node to filter on chunk kind and type
+- IMPROVEMENT: disable endpointing in "deepgram" node as it makes no sense for us
+- IMPROVEMENT: allow -d (dashboard) option to accept a name which is displayed below the dashboard columns
+- IMPROVEMENT: display names below the dashboard columns in UI
+- IMPROVEMENT: move text columns in dashboard UI always to the bottom of the screen
+- IMPROVEMENT: in dashboard UI adjust all colors to be blue as the main theme and make font sizes more eligible
+- BUGFIX: switch back "deepgram" node to use "nova-2" model (as it supports numerals feature, etc)
+- CLEANUP: in "deepgram" node remove ancient initialization workaround
+- CLEANUP: simplify codebase structure and remove redundancy
+- UPGRADE: upgrade NPM dependencies
+
 1.3.2 (2025-08-04)
 ------------------
 

package/etc/stx.conf

@@ -4,74 +4,70 @@
 ##  Licensed under GPL 3.0 <https://spdx.org/licenses/GPL-3.0-only>
 ##
 
-#   make patches
-patch-make
-    npm shrinkwrap && \
-    patch-package --patch-dir package.d "@ericedouard/vad-node-realtime" && \
-    shx rm -f npm-shrinkwrap.json
+#   (internal): just pass-through "npm install" operation
+post-install
+    npm --prefix speechflow-cli   install && \
+    npm --prefix speechflow-ui-db install && \
+    npm --prefix speechflow-ui-st install
 
-#   [INTERNAL] apply patches
-patch-apply
-    patch-package --patch-dir package.d
+#   automatically update all NPM dependencies
+upd
+    npx -y upd && \
+    (cd speechflow-cli && npx -y upd) && \
+    (cd speechflow-ui-db && npx -y upd) && \
+    (cd speechflow-ui-st && npx -y upd)
 
-#   multiview-style development dashboard
-dev
-    stmux -w always -m beep -e "built.in.+ms" -- \
-    [ -s 35% "stx lint-watch" : \
-    -s 15% "stx build-watch" : \
-    -s 30% "stx server-delay server-watch" ]
+#   lint components for development
+lint
+    npm --prefix speechflow-cli   start lint && \
+    npm --prefix speechflow-ui-db start lint && \
+    npm --prefix speechflow-ui-st start lint
 
-#   static code analysis (linting)
-lint:
-    tsc --project etc/tsconfig.json --noEmit && \
-    oxlint --config etc/oxlint.jsonc src/**/*.ts && \
-    biome lint --diagnostic-level=warn --config-path=etc/biome.jsonc src/*.ts && \
-    eslint --config etc/eslint.mjs src/**/*.ts
-
-#   static code analysis (linting) with file watching
-lint-watch
-    nodemon --exec "stx lint" --watch src --ext ts
-
-#   code compilation/transpiling (building)
+#   build components for production (full build)
 build
-    tsc --project etc/tsconfig.json
-
-#   code compilation/transpiling (building) with file watching
-build-watch
-    nodemon --exec "stx build" --watch src --ext ts
+    npm --prefix speechflow-cli   start build && \
+    npm --prefix speechflow-ui-db start build && \
+    npm --prefix speechflow-ui-st start build
 
-#   build all-in-one packages
-build-pkg
-    cd dst && \
-    targets="node24-linux-x64,node24-linux-arm64" && \
-    targets="$targets,node24-win-x64,node24-win-arm64" && \
-    targets="$targets,node24-macos-x64,node24-macos-arm64" && \
-    pkg --sea --public -c ../package.json -t "$targets" speechflow.js && \
-    shx mv speechflow-linux-x64     speechflow-lnx-x64     && \
-    shx mv speechflow-linux-arm64   speechflow-lnx-a64     && \
-    shx mv speechflow-win-x64.exe   speechflow-win-x64.exe && \
-    shx mv speechflow-win-arm64.exe speechflow-win-a64.exe && \
-    shx mv speechflow-macos-x64     speechflow-mac-x64     && \
-    shx mv speechflow-macos-arm64   speechflow-mac-a64
+#   (internal): build components for development (partial build only)
+build:dev
+    npm --prefix speechflow-cli   start build && \
+    npm --prefix speechflow-ui-db start build && \
+    npm --prefix speechflow-ui-st start build
 
-#   [INTERNAL] wait for server
-server-delay
-    delay 2.0
+#   (internal): build components for development (continuous file watching)
+build:dev:watch
+    watch "npm start build:dev" \
+        speechflow-cli/src \
+        speechflow-ui-db/src \
+        speechflow-ui-st/src \
+        --wait=2 --interval=1
 
-#   run program
-server
-    node --enable-source-maps dst/speechflow.js -v info -c studio@etc/speechflow.yaml "$@"
+#   (internal): run command-line interface for production
+run:dev:watch
+    wait-on -l "speechflow-cli/dst/speechflow.js" && \
+    cross-env NODE_OPTIONS="--no-warnings" \
+        nodemon --signal SIGINT -w speechflow-5-cli/dst-stage2 --delay 1500ms \
+            speechflow-cli/dst/speechflow.js -- \
+            $SPEECHFLOW_OPTIONS
 
-#   run program with file watching
-server-watch
-    cross-env NODE_OPTIONS="--enable-source-maps" \
-         nodemon --exec "stx server" --watch dst --ext ts --delay 1.0
+#   (internal): run command-line interface for development (continuous file watching)
+run:dev
+    concurrently -p "[{name}]" -n "build,run" -c "blue,red" \
+        "npm start build:dev:watch" \
+        "npm start run:dev:watch"
 
-#   remove all files regularly generated
+#   remove all generated artifacts (reverse of "npm start build")
 clean
-    shx rm -rf dst
+    npm --prefix speechflow-cli   start clean && \
+    npm --prefix speechflow-ui-db start clean && \
+    npm --prefix speechflow-ui-st start clean
 
-#   remove all files generated
-clean-dist : clean
+#   remove all generated artifacts (reverse of "npm install" and "npm start build")
+clean:dist
+    npm --prefix speechflow-cli   start clean:dist && \
+    npm --prefix speechflow-ui-db start clean:dist && \
+    npm --prefix speechflow-ui-st start clean:dist && \
+    shx rm -f package-lock.json && \
     shx rm -rf node_modules
 

package/package.json

@@ -1,117 +1,36 @@
 {
-    "name":                                 "speechflow",
-    "version":                              "1.3.2",
-    "x-stdver":                             "1.3.2-GA",
-    "x-release":                            "2025-08-04",
-    "homepage":                             "https://github.com/rse/speechflow",
-    "description":                          "Speech Processing Flow Graph",
-    "license":                              "GPL-3.0-only",
+    "name":             "speechflow",
+    "version":          "1.4.0",
+    "x-stdver":         "1.4.0-GA",
+    "x-release":        "2025-08-05",
+    "homepage":         "https://github.com/rse/speechflow",
+    "description":      "Speech Processing Flow Graph",
+    "license":          "GPL-3.0-only",
     "author": {
-        "name":                             "Dr. Ralf S. Engelschall",
-        "email":                            "rse@engelschall.com",
-        "url":                              "http://engelschall.com"
+        "name":         "Dr. Ralf S. Engelschall",
+        "email":        "rse@engelschall.com",
+        "url":          "http://engelschall.com"
     },
     "repository": {
-        "type":                             "git",
-        "url":                              "git+https://github.com/rse/speechflow.git"
-    },
-    "dependencies": {
-        "cli-io":                           "0.9.13",
-        "yargs":                            "18.0.0",
-        "flowlink":                         "1.0.0",
-        "js-yaml":                          "4.1.0",
-        "@gpeng/naudiodon":                 "2.4.1",
-        "@deepgram/sdk":                    "4.11.1",
-        "deepl-node":                       "1.19.0",
-        "@elevenlabs/elevenlabs-js":        "2.7.0",
-        "stream-transform":                 "3.4.0",
-        "get-stream":                       "9.0.1",
-        "@dotenvx/dotenvx":                 "1.48.4",
-        "speex-resampler":                  "3.0.1",
-        "object-path":                      "0.11.8",
-        "ws":                               "8.18.3",
-        "bufferutil":                       "4.0.9",
-        "utf-8-validate":                   "6.0.5",
-        "@hapi/hapi":                       "21.4.0",
-        "@hapi/boom":                       "10.0.1",
-        "@hapi/inert":                      "7.1.0",
-        "hapi-plugin-header":               "1.1.8",
-        "hapi-plugin-websocket":            "2.4.11",
-        "@opensumi/reconnecting-websocket": "4.4.0",
-        "ollama":                           "0.5.16",
-        "openai":                           "5.11.0",
-        "@rse/ffmpeg":                      "1.4.2",
-        "ffmpeg-stream":                    "1.0.1",
-        "installed-packages":               "1.0.13",
-        "syspath":                          "1.0.8",
-        "wav":                              "1.0.2",
-        "mqtt":                             "5.14.0",
-        "cbor2":                            "2.0.1",
-        "arktype":                          "2.1.20",
-        "pure-uuid":                        "1.8.1",
-        "wavefile":                         "11.0.0",
-        "audio-inspect":                    "0.0.3",
-        "@huggingface/transformers":        "3.7.1",
-        "kokoro-js":                        "1.2.1",
-        "@ericedouard/vad-node-realtime":   "0.2.0",
-        "luxon":                            "3.7.1",
-        "node-interval-tree":               "2.1.2",
-        "wrap-text":                        "1.0.10",
-        "cli-table3":                       "0.6.5"
+        "type":         "git",
+        "url":          "git+https://github.com/rse/speechflow.git"
     },
     "devDependencies": {
-        "eslint":                           "9.32.0",
-        "@eslint/js":                       "9.32.0",
-        "neostandard":                      "0.12.2",
-        "eslint-plugin-promise":            "7.2.1",
-        "eslint-plugin-import":             "2.32.0",
-        "eslint-plugin-node":               "11.1.0",
-        "@typescript-eslint/eslint-plugin": "8.38.0",
-        "@typescript-eslint/parser":        "8.38.0",
-        "oxlint":                           "1.9.0",
-        "eslint-plugin-oxlint":             "1.9.0",
-        "@biomejs/biome":                   "2.0.6",
-        "eslint-config-biome":              "2.1.3",
-
-        "@types/node":                      "24.1.0",
-        "@types/yargs":                     "17.0.33",
-        "@types/js-yaml":                   "4.0.9",
-        "@types/object-path":               "0.11.4",
-        "@types/ws":                        "8.18.1",
-        "@types/resolve":                   "1.20.6",
-        "@types/wav":                       "1.0.4",
-        "@types/luxon":                     "3.7.1",
-        "@types/wrap-text":                 "1.0.2",
-
-        "patch-package":                    "8.0.0",
-        "stmux":                            "1.8.11",
-        "nodemon":                          "3.1.10",
-        "shx":                              "0.4.0",
-        "@yao-pkg/pkg":                     "6.6.0",
-        "typescript":                       "5.8.3",
-        "delay-cli":                        "2.0.0",
-        "@rse/stx":                         "1.0.7",
-        "cross-env":                        "10.0.0"
-    },
-    "overrides": {
-        "@huggingface/transformers":        { "onnxruntime-node": "1.23.0-dev.20250703-7fc6235861" }
-    },
-    "upd":                                  [ "!@biomejs/biome", "!typescript" ],
-    "engines": {
-        "node":                             ">=22.0.0"
+        "@rse/stx":     "1.0.7",
+        "nodemon":      "3.1.10",
+        "watch":        "1.0.2",
+        "concurrently": "9.2.0",
+        "wait-on":      "8.0.4",
+        "cross-env":    "10.0.0",
+        "shx":          "0.4.0"
     },
-    "bin":                                  { "speechflow": "dst/speechflow.js" },
-    "types":                                "./dst/speechflow-node.d.ts",
-    "module":                               "./dst/speechflow-node.js",
-    "main":                                 "./dst/speechflow-node.js",
-    "exports": {
-        ".": {
-            "import":                       { "types": "./dst/speechflow-node.d.ts", "default": "./dst/speechflow-node.js" },
-            "require":                      { "types": "./dst/speechflow-node.d.ts", "default": "./dst/speechflow-node.js" }
-        }
+    "engines" : {
+        "npm":          ">=10.0.0",
+        "node":         ">=22.0.0"
     },
     "scripts": {
-        "postinstall":                      "npm start patch-apply",
-        "start":                            "stx -v4 -c etc/stx.conf"
+        "postinstall":  "npm start post-install",
+        "start":        "stx -v4 -c etc/stx.conf",
+        "speechflow":   "node --no-warnings speechflow-cli/dst/speechflow.js"
     }
 }

package/speechflow-cli/etc/stx.conf

@@ -0,0 +1,77 @@
+##
+##  SpeechFlow - Speech Processing Flow Graph
+##  Copyright (c) 2024-2025 Dr. Ralf S. Engelschall <rse@engelschall.com>
+##  Licensed under GPL 3.0 <https://spdx.org/licenses/GPL-3.0-only>
+##
+
+#   make patches
+patch-make
+    npm shrinkwrap && \
+    patch-package --patch-dir package.d "@ericedouard/vad-node-realtime" && \
+    shx rm -f npm-shrinkwrap.json
+
+#   [INTERNAL] apply patches
+patch-apply
+    patch-package --patch-dir package.d
+
+#   multiview-style development dashboard
+dev
+    stmux -w always -m beep -e "built.in.+ms" -- \
+    [ -s 35% "stx lint-watch" : \
+    -s 15% "stx build-watch" : \
+    -s 30% "stx server-delay server-watch" ]
+
+#   static code analysis (linting)
+lint:
+    tsc --project etc/tsconfig.json --noEmit && \
+    oxlint --config etc/oxlint.jsonc src/**/*.ts && \
+    biome lint --diagnostic-level=warn --config-path=etc/biome.jsonc src/*.ts && \
+    eslint --config etc/eslint.mjs src/**/*.ts
+
+#   static code analysis (linting) with file watching
+lint-watch
+    nodemon --exec "stx lint" --watch src --ext ts
+
+#   code compilation/transpiling (building)
+build
+    tsc --project etc/tsconfig.json
+
+#   code compilation/transpiling (building) with file watching
+build-watch
+    nodemon --exec "stx build" --watch src --ext ts
+
+#   build all-in-one packages
+build-pkg
+    cd dst && \
+    targets="node24-linux-x64,node24-linux-arm64" && \
+    targets="$targets,node24-win-x64,node24-win-arm64" && \
+    targets="$targets,node24-macos-x64,node24-macos-arm64" && \
+    pkg --sea --public -c ../package.json -t "$targets" speechflow.js && \
+    shx mv speechflow-linux-x64     speechflow-lnx-x64     && \
+    shx mv speechflow-linux-arm64   speechflow-lnx-a64     && \
+    shx mv speechflow-win-x64.exe   speechflow-win-x64.exe && \
+    shx mv speechflow-win-arm64.exe speechflow-win-a64.exe && \
+    shx mv speechflow-macos-x64     speechflow-mac-x64     && \
+    shx mv speechflow-macos-arm64   speechflow-mac-a64
+
+#   [INTERNAL] wait for server
+server-delay
+    delay 2.0
+
+#   run program
+server
+    node --enable-source-maps dst/speechflow.js -v info -c studio@etc/speechflow.yaml "$@"
+
+#   run program with file watching
+server-watch
+    cross-env NODE_OPTIONS="--enable-source-maps" \
+         nodemon --exec "stx server" --watch dst --ext ts --delay 1.0
+
+#   remove all files regularly generated
+clean
+    shx rm -rf dst
+
+#   remove all files generated
+clean-dist : clean
+    shx rm -rf node_modules
+

package/speechflow-cli/package.json

@@ -0,0 +1,116 @@
+{
+    "private":                              "true",
+    "name":                                 "speechflow",
+    "version":                              "0.0.0",
+    "homepage":                             "https://github.com/rse/speechflow",
+    "description":                          "Speech Processing Flow Graph",
+    "license":                              "GPL-3.0-only",
+    "author": {
+        "name":                             "Dr. Ralf S. Engelschall",
+        "email":                            "rse@engelschall.com",
+        "url":                              "http://engelschall.com"
+    },
+    "repository": {
+        "type":                             "git",
+        "url":                              "git+https://github.com/rse/speechflow.git"
+    },
+    "dependencies": {
+        "cli-io":                           "0.9.13",
+        "yargs":                            "18.0.0",
+        "flowlink":                         "1.0.0",
+        "js-yaml":                          "4.1.0",
+        "@gpeng/naudiodon":                 "2.4.1",
+        "@deepgram/sdk":                    "4.11.1",
+        "deepl-node":                       "1.19.0",
+        "@elevenlabs/elevenlabs-js":        "2.7.0",
+        "stream-transform":                 "3.4.0",
+        "get-stream":                       "9.0.1",
+        "@dotenvx/dotenvx":                 "1.48.4",
+        "speex-resampler":                  "3.0.1",
+        "object-path":                      "0.11.8",
+        "ws":                               "8.18.3",
+        "bufferutil":                       "4.0.9",
+        "utf-8-validate":                   "6.0.5",
+        "@hapi/hapi":                       "21.4.2",
+        "@hapi/boom":                       "10.0.1",
+        "@hapi/inert":                      "7.1.0",
+        "hapi-plugin-header":               "1.1.8",
+        "hapi-plugin-websocket":            "2.4.11",
+        "@opensumi/reconnecting-websocket": "4.4.0",
+        "ollama":                           "0.5.16",
+        "openai":                           "5.12.0",
+        "@rse/ffmpeg":                      "1.4.2",
+        "ffmpeg-stream":                    "1.0.1",
+        "installed-packages":               "1.0.13",
+        "syspath":                          "1.0.8",
+        "wav":                              "1.0.2",
+        "mqtt":                             "5.14.0",
+        "cbor2":                            "2.0.1",
+        "arktype":                          "2.1.20",
+        "pure-uuid":                        "1.8.1",
+        "wavefile":                         "11.0.0",
+        "audio-inspect":                    "0.0.3",
+        "@huggingface/transformers":        "3.7.1",
+        "kokoro-js":                        "1.2.1",
+        "@ericedouard/vad-node-realtime":   "0.2.0",
+        "luxon":                            "3.7.1",
+        "node-interval-tree":               "2.1.2",
+        "wrap-text":                        "1.0.10",
+        "cli-table3":                       "0.6.5"
+    },
+    "devDependencies": {
+        "eslint":                           "9.32.0",
+        "@eslint/js":                       "9.32.0",
+        "neostandard":                      "0.12.2",
+        "eslint-plugin-promise":            "7.2.1",
+        "eslint-plugin-import":             "2.32.0",
+        "eslint-plugin-node":               "11.1.0",
+        "@typescript-eslint/eslint-plugin": "8.39.0",
+        "@typescript-eslint/parser":        "8.39.0",
+        "oxlint":                           "1.9.0",
+        "eslint-plugin-oxlint":             "1.9.0",
+        "@biomejs/biome":                   "2.0.6",
+        "eslint-config-biome":              "2.1.3",
+
+        "@types/node":                      "24.2.0",
+        "@types/yargs":                     "17.0.33",
+        "@types/js-yaml":                   "4.0.9",
+        "@types/object-path":               "0.11.4",
+        "@types/ws":                        "8.18.1",
+        "@types/resolve":                   "1.20.6",
+        "@types/wav":                       "1.0.4",
+        "@types/luxon":                     "3.7.1",
+        "@types/wrap-text":                 "1.0.2",
+
+        "patch-package":                    "8.0.0",
+        "stmux":                            "1.8.11",
+        "nodemon":                          "3.1.10",
+        "shx":                              "0.4.0",
+        "@yao-pkg/pkg":                     "6.6.0",
+        "typescript":                       "5.9.2",
+        "delay-cli":                        "2.0.0",
+        "@rse/stx":                         "1.0.7",
+        "cross-env":                        "10.0.0"
+    },
+    "overrides": {
+        "@huggingface/transformers":        { "onnxruntime-node": "1.23.0-dev.20250703-7fc6235861" }
+    },
+    "upd":                                  [ "!@biomejs/biome" ],
+    "engines": {
+        "node":                             ">=22.0.0"
+    },
+    "bin":                                  { "speechflow": "dst/speechflow.js" },
+    "types":                                "./dst/speechflow-node.d.ts",
+    "module":                               "./dst/speechflow-node.js",
+    "main":                                 "./dst/speechflow-node.js",
+    "exports": {
+        ".": {
+            "import":                       { "types": "./dst/speechflow-node.d.ts", "default": "./dst/speechflow-node.js" },
+            "require":                      { "types": "./dst/speechflow-node.d.ts", "default": "./dst/speechflow-node.js" }
+        }
+    },
+    "scripts": {
+        "postinstall":                      "npm start patch-apply",
+        "start":                            "stx -v4 -c etc/stx.conf"
+    }
+}