Compare commits

..

14 Commits
0.6.1 ... 0.7.3

Author SHA1 Message Date
a07f069ad0 One embedding at a time
All checks were successful
Publish Library / Build NPM Project (push) Successful in 27s
Publish Library / Tag Version (push) Successful in 7s
2026-02-19 22:58:53 -05:00
da15d299e6 parallel embedding cap
All checks were successful
Publish Library / Build NPM Project (push) Successful in 31s
Publish Library / Tag Version (push) Successful in 5s
2026-02-19 21:37:58 -05:00
7ef7c3f676 Cap speaker ID transcript length to 2000 tokens
All checks were successful
Publish Library / Build NPM Project (push) Successful in 34s
Publish Library / Tag Version (push) Successful in 6s
2026-02-14 09:48:12 -05:00
4143d00de7 Working speaker detection with advanced LLM identifying. Improved LLM json function
All checks were successful
Publish Library / Build NPM Project (push) Successful in 39s
Publish Library / Tag Version (push) Successful in 5s
2026-02-14 09:39:17 -05:00
0360f2493d Added hugging face token
All checks were successful
Publish Library / Build NPM Project (push) Successful in 31s
Publish Library / Tag Version (push) Successful in 5s
2026-02-12 22:15:57 -05:00
0172887877 audio worker fix
All checks were successful
Publish Library / Build NPM Project (push) Successful in 28s
Publish Library / Tag Version (push) Successful in 5s
2026-02-12 20:24:12 -05:00
8f89f5e3cf embedding worker fix
All checks were successful
Publish Library / Build NPM Project (push) Successful in 28s
Publish Library / Tag Version (push) Successful in 5s
2026-02-12 20:18:56 -05:00
5bd41f8c6a worker fix?
All checks were successful
Publish Library / Build NPM Project (push) Successful in 29s
Publish Library / Tag Version (push) Successful in 5s
2026-02-12 20:17:31 -05:00
e4399e1b7b Updataes?
All checks were successful
Publish Library / Build NPM Project (push) Successful in 26s
Publish Library / Tag Version (push) Successful in 5s
2026-02-12 20:14:00 -05:00
ad1ee48763 Use one-off workers to process requests without blocking
All checks were successful
Publish Library / Build NPM Project (push) Successful in 27s
Publish Library / Tag Version (push) Successful in 5s
2026-02-12 19:45:17 -05:00
3ed206923f Fix ASR
All checks were successful
Publish Library / Build NPM Project (push) Successful in 27s
Publish Library / Tag Version (push) Successful in 5s
2026-02-12 18:32:19 -05:00
22d5427e86 Fix ASR
All checks were successful
Publish Library / Build NPM Project (push) Successful in 28s
Publish Library / Tag Version (push) Successful in 5s
2026-02-12 17:49:33 -05:00
43b53164c0 Bump 0.6.3
All checks were successful
Publish Library / Build NPM Project (push) Successful in 29s
Publish Library / Tag Version (push) Successful in 4s
2026-02-12 17:24:15 -05:00
575fbac099 Fixed ASR
All checks were successful
Publish Library / Build NPM Project (push) Successful in 30s
Publish Library / Tag Version (push) Successful in 4s
2026-02-12 13:31:30 -05:00
10 changed files with 409 additions and 309 deletions

View File

@@ -75,6 +75,7 @@ A TypeScript library that provides a unified interface for working with multiple
#### Instructions
1. Install the package: `npm i @ztimson/ai-utils`
2. For speaker diarization: `pip install pyannote.audio`
</details>
@@ -90,8 +91,9 @@ A TypeScript library that provides a unified interface for working with multiple
#### Instructions
1. Install the dependencies: `npm i`
2. Build library: `npm build`
3. Run unit tests: `npm test`
2. For speaker diarization: `pip install pyannote.audio`
3. Build library: `npm build`
4. Run unit tests: `npm test`
</details>

337
package-lock.json generated
View File

@@ -1,12 +1,12 @@
{
"name": "@ztimson/ai-utils",
"version": "0.4.0",
"version": "0.6.4",
"lockfileVersion": 3,
"requires": true,
"packages": {
"": {
"name": "@ztimson/ai-utils",
"version": "0.4.0",
"version": "0.6.4",
"license": "MIT",
"dependencies": {
"@anthropic-ai/sdk": "^0.67.0",
@@ -16,7 +16,8 @@
"@ztimson/utils": "^0.27.9",
"cheerio": "^1.2.0",
"openai": "^6.6.0",
"tesseract.js": "^6.0.1"
"tesseract.js": "^6.0.1",
"wavefile": "^11.0.0"
},
"devDependencies": {
"@types/node": "^24.8.1",
@@ -106,9 +107,9 @@
}
},
"node_modules/@esbuild/aix-ppc64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/aix-ppc64/-/aix-ppc64-0.27.2.tgz",
"integrity": "sha512-GZMB+a0mOMZs4MpDbj8RJp4cw+w1WV5NYD6xzgvzUJ5Ek2jerwfO2eADyI6ExDSUED+1X8aMbegahsJi+8mgpw==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/aix-ppc64/-/aix-ppc64-0.27.3.tgz",
"integrity": "sha512-9fJMTNFTWZMh5qwrBItuziu834eOCUcEqymSH7pY+zoMVEZg3gcPuBNxH1EvfVYe9h0x/Ptw8KBzv7qxb7l8dg==",
"cpu": [
"ppc64"
],
@@ -123,9 +124,9 @@
}
},
"node_modules/@esbuild/android-arm": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/android-arm/-/android-arm-0.27.2.tgz",
"integrity": "sha512-DVNI8jlPa7Ujbr1yjU2PfUSRtAUZPG9I1RwW4F4xFB1Imiu2on0ADiI/c3td+KmDtVKNbi+nffGDQMfcIMkwIA==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/android-arm/-/android-arm-0.27.3.tgz",
"integrity": "sha512-i5D1hPY7GIQmXlXhs2w8AWHhenb00+GxjxRncS2ZM7YNVGNfaMxgzSGuO8o8SJzRc/oZwU2bcScvVERk03QhzA==",
"cpu": [
"arm"
],
@@ -140,9 +141,9 @@
}
},
"node_modules/@esbuild/android-arm64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/android-arm64/-/android-arm64-0.27.2.tgz",
"integrity": "sha512-pvz8ZZ7ot/RBphf8fv60ljmaoydPU12VuXHImtAs0XhLLw+EXBi2BLe3OYSBslR4rryHvweW5gmkKFwTiFy6KA==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/android-arm64/-/android-arm64-0.27.3.tgz",
"integrity": "sha512-YdghPYUmj/FX2SYKJ0OZxf+iaKgMsKHVPF1MAq/P8WirnSpCStzKJFjOjzsW0QQ7oIAiccHdcqjbHmJxRb/dmg==",
"cpu": [
"arm64"
],
@@ -157,9 +158,9 @@
}
},
"node_modules/@esbuild/android-x64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/android-x64/-/android-x64-0.27.2.tgz",
"integrity": "sha512-z8Ank4Byh4TJJOh4wpz8g2vDy75zFL0TlZlkUkEwYXuPSgX8yzep596n6mT7905kA9uHZsf/o2OJZubl2l3M7A==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/android-x64/-/android-x64-0.27.3.tgz",
"integrity": "sha512-IN/0BNTkHtk8lkOM8JWAYFg4ORxBkZQf9zXiEOfERX/CzxW3Vg1ewAhU7QSWQpVIzTW+b8Xy+lGzdYXV6UZObQ==",
"cpu": [
"x64"
],
@@ -174,9 +175,9 @@
}
},
"node_modules/@esbuild/darwin-arm64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/darwin-arm64/-/darwin-arm64-0.27.2.tgz",
"integrity": "sha512-davCD2Zc80nzDVRwXTcQP/28fiJbcOwvdolL0sOiOsbwBa72kegmVU0Wrh1MYrbuCL98Omp5dVhQFWRKR2ZAlg==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/darwin-arm64/-/darwin-arm64-0.27.3.tgz",
"integrity": "sha512-Re491k7ByTVRy0t3EKWajdLIr0gz2kKKfzafkth4Q8A5n1xTHrkqZgLLjFEHVD+AXdUGgQMq+Godfq45mGpCKg==",
"cpu": [
"arm64"
],
@@ -191,9 +192,9 @@
}
},
"node_modules/@esbuild/darwin-x64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/darwin-x64/-/darwin-x64-0.27.2.tgz",
"integrity": "sha512-ZxtijOmlQCBWGwbVmwOF/UCzuGIbUkqB1faQRf5akQmxRJ1ujusWsb3CVfk/9iZKr2L5SMU5wPBi1UWbvL+VQA==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/darwin-x64/-/darwin-x64-0.27.3.tgz",
"integrity": "sha512-vHk/hA7/1AckjGzRqi6wbo+jaShzRowYip6rt6q7VYEDX4LEy1pZfDpdxCBnGtl+A5zq8iXDcyuxwtv3hNtHFg==",
"cpu": [
"x64"
],
@@ -208,9 +209,9 @@
}
},
"node_modules/@esbuild/freebsd-arm64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/freebsd-arm64/-/freebsd-arm64-0.27.2.tgz",
"integrity": "sha512-lS/9CN+rgqQ9czogxlMcBMGd+l8Q3Nj1MFQwBZJyoEKI50XGxwuzznYdwcav6lpOGv5BqaZXqvBSiB/kJ5op+g==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/freebsd-arm64/-/freebsd-arm64-0.27.3.tgz",
"integrity": "sha512-ipTYM2fjt3kQAYOvo6vcxJx3nBYAzPjgTCk7QEgZG8AUO3ydUhvelmhrbOheMnGOlaSFUoHXB6un+A7q4ygY9w==",
"cpu": [
"arm64"
],
@@ -225,9 +226,9 @@
}
},
"node_modules/@esbuild/freebsd-x64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/freebsd-x64/-/freebsd-x64-0.27.2.tgz",
"integrity": "sha512-tAfqtNYb4YgPnJlEFu4c212HYjQWSO/w/h/lQaBK7RbwGIkBOuNKQI9tqWzx7Wtp7bTPaGC6MJvWI608P3wXYA==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/freebsd-x64/-/freebsd-x64-0.27.3.tgz",
"integrity": "sha512-dDk0X87T7mI6U3K9VjWtHOXqwAMJBNN2r7bejDsc+j03SEjtD9HrOl8gVFByeM0aJksoUuUVU9TBaZa2rgj0oA==",
"cpu": [
"x64"
],
@@ -242,9 +243,9 @@
}
},
"node_modules/@esbuild/linux-arm": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/linux-arm/-/linux-arm-0.27.2.tgz",
"integrity": "sha512-vWfq4GaIMP9AIe4yj1ZUW18RDhx6EPQKjwe7n8BbIecFtCQG4CfHGaHuh7fdfq+y3LIA2vGS/o9ZBGVxIDi9hw==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/linux-arm/-/linux-arm-0.27.3.tgz",
"integrity": "sha512-s6nPv2QkSupJwLYyfS+gwdirm0ukyTFNl3KTgZEAiJDd+iHZcbTPPcWCcRYH+WlNbwChgH2QkE9NSlNrMT8Gfw==",
"cpu": [
"arm"
],
@@ -259,9 +260,9 @@
}
},
"node_modules/@esbuild/linux-arm64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/linux-arm64/-/linux-arm64-0.27.2.tgz",
"integrity": "sha512-hYxN8pr66NsCCiRFkHUAsxylNOcAQaxSSkHMMjcpx0si13t1LHFphxJZUiGwojB1a/Hd5OiPIqDdXONia6bhTw==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/linux-arm64/-/linux-arm64-0.27.3.tgz",
"integrity": "sha512-sZOuFz/xWnZ4KH3YfFrKCf1WyPZHakVzTiqji3WDc0BCl2kBwiJLCXpzLzUBLgmp4veFZdvN5ChW4Eq/8Fc2Fg==",
"cpu": [
"arm64"
],
@@ -276,9 +277,9 @@
}
},
"node_modules/@esbuild/linux-ia32": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/linux-ia32/-/linux-ia32-0.27.2.tgz",
"integrity": "sha512-MJt5BRRSScPDwG2hLelYhAAKh9imjHK5+NE/tvnRLbIqUWa+0E9N4WNMjmp/kXXPHZGqPLxggwVhz7QP8CTR8w==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/linux-ia32/-/linux-ia32-0.27.3.tgz",
"integrity": "sha512-yGlQYjdxtLdh0a3jHjuwOrxQjOZYD/C9PfdbgJJF3TIZWnm/tMd/RcNiLngiu4iwcBAOezdnSLAwQDPqTmtTYg==",
"cpu": [
"ia32"
],
@@ -293,9 +294,9 @@
}
},
"node_modules/@esbuild/linux-loong64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/linux-loong64/-/linux-loong64-0.27.2.tgz",
"integrity": "sha512-lugyF1atnAT463aO6KPshVCJK5NgRnU4yb3FUumyVz+cGvZbontBgzeGFO1nF+dPueHD367a2ZXe1NtUkAjOtg==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/linux-loong64/-/linux-loong64-0.27.3.tgz",
"integrity": "sha512-WO60Sn8ly3gtzhyjATDgieJNet/KqsDlX5nRC5Y3oTFcS1l0KWba+SEa9Ja1GfDqSF1z6hif/SkpQJbL63cgOA==",
"cpu": [
"loong64"
],
@@ -310,9 +311,9 @@
}
},
"node_modules/@esbuild/linux-mips64el": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/linux-mips64el/-/linux-mips64el-0.27.2.tgz",
"integrity": "sha512-nlP2I6ArEBewvJ2gjrrkESEZkB5mIoaTswuqNFRv/WYd+ATtUpe9Y09RnJvgvdag7he0OWgEZWhviS1OTOKixw==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/linux-mips64el/-/linux-mips64el-0.27.3.tgz",
"integrity": "sha512-APsymYA6sGcZ4pD6k+UxbDjOFSvPWyZhjaiPyl/f79xKxwTnrn5QUnXR5prvetuaSMsb4jgeHewIDCIWljrSxw==",
"cpu": [
"mips64el"
],
@@ -327,9 +328,9 @@
}
},
"node_modules/@esbuild/linux-ppc64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/linux-ppc64/-/linux-ppc64-0.27.2.tgz",
"integrity": "sha512-C92gnpey7tUQONqg1n6dKVbx3vphKtTHJaNG2Ok9lGwbZil6DrfyecMsp9CrmXGQJmZ7iiVXvvZH6Ml5hL6XdQ==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/linux-ppc64/-/linux-ppc64-0.27.3.tgz",
"integrity": "sha512-eizBnTeBefojtDb9nSh4vvVQ3V9Qf9Df01PfawPcRzJH4gFSgrObw+LveUyDoKU3kxi5+9RJTCWlj4FjYXVPEA==",
"cpu": [
"ppc64"
],
@@ -344,9 +345,9 @@
}
},
"node_modules/@esbuild/linux-riscv64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/linux-riscv64/-/linux-riscv64-0.27.2.tgz",
"integrity": "sha512-B5BOmojNtUyN8AXlK0QJyvjEZkWwy/FKvakkTDCziX95AowLZKR6aCDhG7LeF7uMCXEJqwa8Bejz5LTPYm8AvA==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/linux-riscv64/-/linux-riscv64-0.27.3.tgz",
"integrity": "sha512-3Emwh0r5wmfm3ssTWRQSyVhbOHvqegUDRd0WhmXKX2mkHJe1SFCMJhagUleMq+Uci34wLSipf8Lagt4LlpRFWQ==",
"cpu": [
"riscv64"
],
@@ -361,9 +362,9 @@
}
},
"node_modules/@esbuild/linux-s390x": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/linux-s390x/-/linux-s390x-0.27.2.tgz",
"integrity": "sha512-p4bm9+wsPwup5Z8f4EpfN63qNagQ47Ua2znaqGH6bqLlmJ4bx97Y9JdqxgGZ6Y8xVTixUnEkoKSHcpRlDnNr5w==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/linux-s390x/-/linux-s390x-0.27.3.tgz",
"integrity": "sha512-pBHUx9LzXWBc7MFIEEL0yD/ZVtNgLytvx60gES28GcWMqil8ElCYR4kvbV2BDqsHOvVDRrOxGySBM9Fcv744hw==",
"cpu": [
"s390x"
],
@@ -378,9 +379,9 @@
}
},
"node_modules/@esbuild/linux-x64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/linux-x64/-/linux-x64-0.27.2.tgz",
"integrity": "sha512-uwp2Tip5aPmH+NRUwTcfLb+W32WXjpFejTIOWZFw/v7/KnpCDKG66u4DLcurQpiYTiYwQ9B7KOeMJvLCu/OvbA==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/linux-x64/-/linux-x64-0.27.3.tgz",
"integrity": "sha512-Czi8yzXUWIQYAtL/2y6vogER8pvcsOsk5cpwL4Gk5nJqH5UZiVByIY8Eorm5R13gq+DQKYg0+JyQoytLQas4dA==",
"cpu": [
"x64"
],
@@ -395,9 +396,9 @@
}
},
"node_modules/@esbuild/netbsd-arm64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/netbsd-arm64/-/netbsd-arm64-0.27.2.tgz",
"integrity": "sha512-Kj6DiBlwXrPsCRDeRvGAUb/LNrBASrfqAIok+xB0LxK8CHqxZ037viF13ugfsIpePH93mX7xfJp97cyDuTZ3cw==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/netbsd-arm64/-/netbsd-arm64-0.27.3.tgz",
"integrity": "sha512-sDpk0RgmTCR/5HguIZa9n9u+HVKf40fbEUt+iTzSnCaGvY9kFP0YKBWZtJaraonFnqef5SlJ8/TiPAxzyS+UoA==",
"cpu": [
"arm64"
],
@@ -412,9 +413,9 @@
}
},
"node_modules/@esbuild/netbsd-x64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/netbsd-x64/-/netbsd-x64-0.27.2.tgz",
"integrity": "sha512-HwGDZ0VLVBY3Y+Nw0JexZy9o/nUAWq9MlV7cahpaXKW6TOzfVno3y3/M8Ga8u8Yr7GldLOov27xiCnqRZf0tCA==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/netbsd-x64/-/netbsd-x64-0.27.3.tgz",
"integrity": "sha512-P14lFKJl/DdaE00LItAukUdZO5iqNH7+PjoBm+fLQjtxfcfFE20Xf5CrLsmZdq5LFFZzb5JMZ9grUwvtVYzjiA==",
"cpu": [
"x64"
],
@@ -429,9 +430,9 @@
}
},
"node_modules/@esbuild/openbsd-arm64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/openbsd-arm64/-/openbsd-arm64-0.27.2.tgz",
"integrity": "sha512-DNIHH2BPQ5551A7oSHD0CKbwIA/Ox7+78/AWkbS5QoRzaqlev2uFayfSxq68EkonB+IKjiuxBFoV8ESJy8bOHA==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/openbsd-arm64/-/openbsd-arm64-0.27.3.tgz",
"integrity": "sha512-AIcMP77AvirGbRl/UZFTq5hjXK+2wC7qFRGoHSDrZ5v5b8DK/GYpXW3CPRL53NkvDqb9D+alBiC/dV0Fb7eJcw==",
"cpu": [
"arm64"
],
@@ -446,9 +447,9 @@
}
},
"node_modules/@esbuild/openbsd-x64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/openbsd-x64/-/openbsd-x64-0.27.2.tgz",
"integrity": "sha512-/it7w9Nb7+0KFIzjalNJVR5bOzA9Vay+yIPLVHfIQYG/j+j9VTH84aNB8ExGKPU4AzfaEvN9/V4HV+F+vo8OEg==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/openbsd-x64/-/openbsd-x64-0.27.3.tgz",
"integrity": "sha512-DnW2sRrBzA+YnE70LKqnM3P+z8vehfJWHXECbwBmH/CU51z6FiqTQTHFenPlHmo3a8UgpLyH3PT+87OViOh1AQ==",
"cpu": [
"x64"
],
@@ -463,9 +464,9 @@
}
},
"node_modules/@esbuild/openharmony-arm64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/openharmony-arm64/-/openharmony-arm64-0.27.2.tgz",
"integrity": "sha512-LRBbCmiU51IXfeXk59csuX/aSaToeG7w48nMwA6049Y4J4+VbWALAuXcs+qcD04rHDuSCSRKdmY63sruDS5qag==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/openharmony-arm64/-/openharmony-arm64-0.27.3.tgz",
"integrity": "sha512-NinAEgr/etERPTsZJ7aEZQvvg/A6IsZG/LgZy+81wON2huV7SrK3e63dU0XhyZP4RKGyTm7aOgmQk0bGp0fy2g==",
"cpu": [
"arm64"
],
@@ -480,9 +481,9 @@
}
},
"node_modules/@esbuild/sunos-x64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/sunos-x64/-/sunos-x64-0.27.2.tgz",
"integrity": "sha512-kMtx1yqJHTmqaqHPAzKCAkDaKsffmXkPHThSfRwZGyuqyIeBvf08KSsYXl+abf5HDAPMJIPnbBfXvP2ZC2TfHg==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/sunos-x64/-/sunos-x64-0.27.3.tgz",
"integrity": "sha512-PanZ+nEz+eWoBJ8/f8HKxTTD172SKwdXebZ0ndd953gt1HRBbhMsaNqjTyYLGLPdoWHy4zLU7bDVJztF5f3BHA==",
"cpu": [
"x64"
],
@@ -497,9 +498,9 @@
}
},
"node_modules/@esbuild/win32-arm64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/win32-arm64/-/win32-arm64-0.27.2.tgz",
"integrity": "sha512-Yaf78O/B3Kkh+nKABUF++bvJv5Ijoy9AN1ww904rOXZFLWVc5OLOfL56W+C8F9xn5JQZa3UX6m+IktJnIb1Jjg==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/win32-arm64/-/win32-arm64-0.27.3.tgz",
"integrity": "sha512-B2t59lWWYrbRDw/tjiWOuzSsFh1Y/E95ofKz7rIVYSQkUYBjfSgf6oeYPNWHToFRr2zx52JKApIcAS/D5TUBnA==",
"cpu": [
"arm64"
],
@@ -514,9 +515,9 @@
}
},
"node_modules/@esbuild/win32-ia32": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/win32-ia32/-/win32-ia32-0.27.2.tgz",
"integrity": "sha512-Iuws0kxo4yusk7sw70Xa2E2imZU5HoixzxfGCdxwBdhiDgt9vX9VUCBhqcwY7/uh//78A1hMkkROMJq9l27oLQ==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/win32-ia32/-/win32-ia32-0.27.3.tgz",
"integrity": "sha512-QLKSFeXNS8+tHW7tZpMtjlNb7HKau0QDpwm49u0vUp9y1WOF+PEzkU84y9GqYaAVW8aH8f3GcBck26jh54cX4Q==",
"cpu": [
"ia32"
],
@@ -531,9 +532,9 @@
}
},
"node_modules/@esbuild/win32-x64": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/@esbuild/win32-x64/-/win32-x64-0.27.2.tgz",
"integrity": "sha512-sRdU18mcKf7F+YgheI/zGf5alZatMUTKj/jNS6l744f9u3WFu4v7twcUI9vu4mknF4Y9aDlblIie0IM+5xxaqQ==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/@esbuild/win32-x64/-/win32-x64-0.27.3.tgz",
"integrity": "sha512-4uJGhsxuptu3OcpVAzli+/gWusVGwZZHTlS63hh++ehExkVT8SgiEf7/uC/PclrPPkLhZqGgCTjd0VWLo6xMqA==",
"cpu": [
"x64"
],
@@ -567,9 +568,9 @@
}
},
"node_modules/@isaacs/brace-expansion": {
"version": "5.0.0",
"resolved": "https://registry.npmjs.org/@isaacs/brace-expansion/-/brace-expansion-5.0.0.tgz",
"integrity": "sha512-ZT55BDLV0yv0RBm2czMiZ+SqCGO7AvmOM3G/w2xhVPH+te0aKgFjmBvGlL1dH+ql2tgGO3MVrbb3jCKyvpgnxA==",
"version": "5.0.1",
"resolved": "https://registry.npmjs.org/@isaacs/brace-expansion/-/brace-expansion-5.0.1.tgz",
"integrity": "sha512-WMz71T1JS624nWj2n2fnYAuPovhv7EUhk69R6i9dsVyzxt5eM3bjwvgk9L+APE1TRscGysAVMANkB0jh0LQZrQ==",
"dev": true,
"license": "MIT",
"dependencies": {
@@ -587,9 +588,9 @@
"license": "MIT"
},
"node_modules/@microsoft/api-extractor": {
"version": "7.56.0",
"resolved": "https://registry.npmjs.org/@microsoft/api-extractor/-/api-extractor-7.56.0.tgz",
"integrity": "sha512-H0V69QG5jIb9Ayx35NVBv2lOgFSS3q+Eab2oyGEy0POL3ovYPST+rCNPbwYoczOZXNG8IKjWUmmAMxmDTsXlQA==",
"version": "7.56.3",
"resolved": "https://registry.npmjs.org/@microsoft/api-extractor/-/api-extractor-7.56.3.tgz",
"integrity": "sha512-fRqok4aRNq5GpgGBv2fKlSSKbirPKTJ75vQefthB5x9dwt4Zz+AezUzdc1p/AG4wUBIgmhjcEwn/Rj+N4Wh4Mw==",
"dev": true,
"license": "MIT",
"dependencies": {
@@ -599,10 +600,10 @@
"@rushstack/node-core-library": "5.19.1",
"@rushstack/rig-package": "0.6.0",
"@rushstack/terminal": "0.21.0",
"@rushstack/ts-command-line": "5.1.7",
"@rushstack/ts-command-line": "5.2.0",
"diff": "~8.0.2",
"lodash": "~4.17.15",
"minimatch": "10.0.3",
"lodash": "~4.17.23",
"minimatch": "10.1.2",
"resolve": "~1.22.1",
"semver": "~7.5.4",
"source-map": "~0.6.1",
@@ -625,13 +626,13 @@
}
},
"node_modules/@microsoft/api-extractor/node_modules/minimatch": {
"version": "10.0.3",
"resolved": "https://registry.npmjs.org/minimatch/-/minimatch-10.0.3.tgz",
"integrity": "sha512-IPZ167aShDZZUMdRk66cyQAW3qr0WzbHkPdMYa8bzZhlHhO3jALbKdxcaak7W9FfT2rZNpQuUu4Od7ILEpXSaw==",
"version": "10.1.2",
"resolved": "https://registry.npmjs.org/minimatch/-/minimatch-10.1.2.tgz",
"integrity": "sha512-fu656aJ0n2kcXwsnwnv9g24tkU5uSmOlTjd6WyyaKm2Z+h1qmY6bAjrcaIxF/BslFqbZ8UBtbJi7KgQOZD2PTw==",
"dev": true,
"license": "ISC",
"license": "BlueOak-1.0.0",
"dependencies": {
"@isaacs/brace-expansion": "^5.0.0"
"@isaacs/brace-expansion": "^5.0.1"
},
"engines": {
"node": "20 || >=22"
@@ -1261,9 +1262,9 @@
}
},
"node_modules/@rushstack/ts-command-line": {
"version": "5.1.7",
"resolved": "https://registry.npmjs.org/@rushstack/ts-command-line/-/ts-command-line-5.1.7.tgz",
"integrity": "sha512-Ugwl6flarZcL2nqH5IXFYk3UR3mBVDsVFlCQW/Oaqidvdb/5Ota6b/Z3JXWIdqV3rOR2/JrYoAHanWF5rgenXA==",
"version": "5.2.0",
"resolved": "https://registry.npmjs.org/@rushstack/ts-command-line/-/ts-command-line-5.2.0.tgz",
"integrity": "sha512-lYxCX0nDdkDtCkVpvF0m25ymf66SaMWuppbD6b7MdkIzvGXKBXNIVZlwBH/C0YfkanrupnICWf2n4z3AKSfaHw==",
"dev": true,
"license": "MIT",
"dependencies": {
@@ -1503,9 +1504,9 @@
}
},
"node_modules/@types/node": {
"version": "24.10.9",
"resolved": "https://registry.npmjs.org/@types/node/-/node-24.10.9.tgz",
"integrity": "sha512-ne4A0IpG3+2ETuREInjPNhUGis1SFjv1d5asp8MzEAGtOZeTeHVDOYqOgqfhvseqg/iXty2hjBf1zAOb7RNiNw==",
"version": "24.10.13",
"resolved": "https://registry.npmjs.org/@types/node/-/node-24.10.13.tgz",
"integrity": "sha512-oH72nZRfDv9lADUBSo104Aq7gPHpQZc4BTx38r9xf9pg5LfP6EzSyH2n7qFmmxRQXh7YlUXODcYsg6PuTDSxGg==",
"license": "MIT",
"dependencies": {
"undici-types": "~7.16.0"
@@ -1577,15 +1578,15 @@
}
},
"node_modules/@vue/compiler-core": {
"version": "3.5.27",
"resolved": "https://registry.npmjs.org/@vue/compiler-core/-/compiler-core-3.5.27.tgz",
"integrity": "sha512-gnSBQjZA+//qDZen+6a2EdHqJ68Z7uybrMf3SPjEGgG4dicklwDVmMC1AeIHxtLVPT7sn6sH1KOO+tS6gwOUeQ==",
"version": "3.5.28",
"resolved": "https://registry.npmjs.org/@vue/compiler-core/-/compiler-core-3.5.28.tgz",
"integrity": "sha512-kviccYxTgoE8n6OCw96BNdYlBg2GOWfBuOW4Vqwrt7mSKWKwFVvI8egdTltqRgITGPsTFYtKYfxIG8ptX2PJHQ==",
"dev": true,
"license": "MIT",
"dependencies": {
"@babel/parser": "^7.28.5",
"@vue/shared": "3.5.27",
"entities": "^7.0.0",
"@babel/parser": "^7.29.0",
"@vue/shared": "3.5.28",
"entities": "^7.0.1",
"estree-walker": "^2.0.2",
"source-map-js": "^1.2.1"
}
@@ -1604,14 +1605,14 @@
}
},
"node_modules/@vue/compiler-dom": {
"version": "3.5.27",
"resolved": "https://registry.npmjs.org/@vue/compiler-dom/-/compiler-dom-3.5.27.tgz",
"integrity": "sha512-oAFea8dZgCtVVVTEC7fv3T5CbZW9BxpFzGGxC79xakTr6ooeEqmRuvQydIiDAkglZEAd09LgVf1RoDnL54fu5w==",
"version": "3.5.28",
"resolved": "https://registry.npmjs.org/@vue/compiler-dom/-/compiler-dom-3.5.28.tgz",
"integrity": "sha512-/1ZepxAb159jKR1btkefDP+J2xuWL5V3WtleRmxaT+K2Aqiek/Ab/+Ebrw2pPj0sdHO8ViAyyJWfhXXOP/+LQA==",
"dev": true,
"license": "MIT",
"dependencies": {
"@vue/compiler-core": "3.5.27",
"@vue/shared": "3.5.27"
"@vue/compiler-core": "3.5.28",
"@vue/shared": "3.5.28"
}
},
"node_modules/@vue/compiler-vue2": {
@@ -1651,9 +1652,9 @@
}
},
"node_modules/@vue/shared": {
"version": "3.5.27",
"resolved": "https://registry.npmjs.org/@vue/shared/-/shared-3.5.27.tgz",
"integrity": "sha512-dXr/3CgqXsJkZ0n9F3I4elY8wM9jMJpP3pvRG52r6m0tu/MsAFIe6JpXVGeNMd/D9F4hQynWT8Rfuj0bdm9kFQ==",
"version": "3.5.28",
"resolved": "https://registry.npmjs.org/@vue/shared/-/shared-3.5.28.tgz",
"integrity": "sha512-cfWa1fCGBxrvaHRhvV3Is0MgmrbSCxYTXCSCau2I0a1Xw1N1pHAvkWCiXPRAqjvToILvguNyEwjevUqAuBQWvQ==",
"dev": true,
"license": "MIT"
},
@@ -1837,9 +1838,9 @@
}
},
"node_modules/bare-fs": {
"version": "4.5.3",
"resolved": "https://registry.npmjs.org/bare-fs/-/bare-fs-4.5.3.tgz",
"integrity": "sha512-9+kwVx8QYvt3hPWnmb19tPnh38c6Nihz8Lx3t0g9+4GoIf3/fTgYwM4Z6NxgI+B9elLQA7mLE9PpqcWtOMRDiQ==",
"version": "4.5.4",
"resolved": "https://registry.npmjs.org/bare-fs/-/bare-fs-4.5.4.tgz",
"integrity": "sha512-POK4oplfA7P7gqvetNmCs4CNtm9fNsx+IAh7jH7GgU0OJdge2rso0R20TNWVq6VoWcCvsTdlNDaleLHGaKx8CA==",
"license": "Apache-2.0",
"optional": true,
"dependencies": {
@@ -2183,9 +2184,9 @@
"license": "MIT"
},
"node_modules/confbox": {
"version": "0.2.2",
"resolved": "https://registry.npmjs.org/confbox/-/confbox-0.2.2.tgz",
"integrity": "sha512-1NB+BKqhtNipMsov4xI/NnhCKp9XG9NamYp5PVm9klAT0fsrNPjaFICsCFhNhwZJKNh7zB/3q8qXz0E9oaMNtQ==",
"version": "0.2.4",
"resolved": "https://registry.npmjs.org/confbox/-/confbox-0.2.4.tgz",
"integrity": "sha512-ysOGlgTFbN2/Y6Cg3Iye8YKulHw+R2fNXHrgSmXISQdMnomY6eNDprVdW9R5xBguEqI954+S6709UyiO7B+6OQ==",
"dev": true,
"license": "MIT"
},
@@ -2491,9 +2492,9 @@
}
},
"node_modules/esbuild": {
"version": "0.27.2",
"resolved": "https://registry.npmjs.org/esbuild/-/esbuild-0.27.2.tgz",
"integrity": "sha512-HyNQImnsOC7X9PMNaCIeAm4ISCQXs5a5YasTXVliKv4uuBo1dKrG0A+uQS8M5eXjVMnLg3WgXaKvprHlFJQffw==",
"version": "0.27.3",
"resolved": "https://registry.npmjs.org/esbuild/-/esbuild-0.27.3.tgz",
"integrity": "sha512-8VwMnyGCONIs6cWue2IdpHxHnAjzxnw2Zr7MkVxB2vjmQ2ivqGFb4LEG3SMnv0Gb2F/G/2yA8zUaiL1gywDCCg==",
"dev": true,
"hasInstallScript": true,
"license": "MIT",
@@ -2504,32 +2505,32 @@
"node": ">=18"
},
"optionalDependencies": {
"@esbuild/aix-ppc64": "0.27.2",
"@esbuild/android-arm": "0.27.2",
"@esbuild/android-arm64": "0.27.2",
"@esbuild/android-x64": "0.27.2",
"@esbuild/darwin-arm64": "0.27.2",
"@esbuild/darwin-x64": "0.27.2",
"@esbuild/freebsd-arm64": "0.27.2",
"@esbuild/freebsd-x64": "0.27.2",
"@esbuild/linux-arm": "0.27.2",
"@esbuild/linux-arm64": "0.27.2",
"@esbuild/linux-ia32": "0.27.2",
"@esbuild/linux-loong64": "0.27.2",
"@esbuild/linux-mips64el": "0.27.2",
"@esbuild/linux-ppc64": "0.27.2",
"@esbuild/linux-riscv64": "0.27.2",
"@esbuild/linux-s390x": "0.27.2",
"@esbuild/linux-x64": "0.27.2",
"@esbuild/netbsd-arm64": "0.27.2",
"@esbuild/netbsd-x64": "0.27.2",
"@esbuild/openbsd-arm64": "0.27.2",
"@esbuild/openbsd-x64": "0.27.2",
"@esbuild/openharmony-arm64": "0.27.2",
"@esbuild/sunos-x64": "0.27.2",
"@esbuild/win32-arm64": "0.27.2",
"@esbuild/win32-ia32": "0.27.2",
"@esbuild/win32-x64": "0.27.2"
"@esbuild/aix-ppc64": "0.27.3",
"@esbuild/android-arm": "0.27.3",
"@esbuild/android-arm64": "0.27.3",
"@esbuild/android-x64": "0.27.3",
"@esbuild/darwin-arm64": "0.27.3",
"@esbuild/darwin-x64": "0.27.3",
"@esbuild/freebsd-arm64": "0.27.3",
"@esbuild/freebsd-x64": "0.27.3",
"@esbuild/linux-arm": "0.27.3",
"@esbuild/linux-arm64": "0.27.3",
"@esbuild/linux-ia32": "0.27.3",
"@esbuild/linux-loong64": "0.27.3",
"@esbuild/linux-mips64el": "0.27.3",
"@esbuild/linux-ppc64": "0.27.3",
"@esbuild/linux-riscv64": "0.27.3",
"@esbuild/linux-s390x": "0.27.3",
"@esbuild/linux-x64": "0.27.3",
"@esbuild/netbsd-arm64": "0.27.3",
"@esbuild/netbsd-x64": "0.27.3",
"@esbuild/openbsd-arm64": "0.27.3",
"@esbuild/openbsd-x64": "0.27.3",
"@esbuild/openharmony-arm64": "0.27.3",
"@esbuild/sunos-x64": "0.27.3",
"@esbuild/win32-arm64": "0.27.3",
"@esbuild/win32-ia32": "0.27.3",
"@esbuild/win32-x64": "0.27.3"
}
},
"node_modules/escalade": {
@@ -3102,9 +3103,9 @@
}
},
"node_modules/markdown-it": {
"version": "14.1.0",
"resolved": "https://registry.npmjs.org/markdown-it/-/markdown-it-14.1.0.tgz",
"integrity": "sha512-a54IwgWPaeBCAAsv13YgmALOF1elABB08FxO9i+r4VFk5Vl4pKokRPeX8u5TCgSsPi6ec1otfLjdOpVcgbpshg==",
"version": "14.1.1",
"resolved": "https://registry.npmjs.org/markdown-it/-/markdown-it-14.1.1.tgz",
"integrity": "sha512-BuU2qnTti9YKgK5N+IeMubp14ZUKUUw7yeJbkjtosvHiP0AZ5c8IAgEMk79D0eC8F23r4Ac/q8cAIFdm2FtyoA==",
"dev": true,
"license": "MIT",
"dependencies": {
@@ -3509,9 +3510,9 @@
}
},
"node_modules/openai": {
"version": "6.17.0",
"resolved": "https://registry.npmjs.org/openai/-/openai-6.17.0.tgz",
"integrity": "sha512-NHRpPEUPzAvFOAFs9+9pC6+HCw/iWsYsKCMPXH5Kw7BpMxqd8g/A07/1o7Gx2TWtCnzevVRyKMRFqyiHyAlqcA==",
"version": "6.21.0",
"resolved": "https://registry.npmjs.org/openai/-/openai-6.21.0.tgz",
"integrity": "sha512-26dQFi76dB8IiN/WKGQOV+yKKTTlRCxQjoi2WLt0kMcH8pvxVyvfdBDkld5GTl7W1qvBpwVOtFcsqktj3fBRpA==",
"license": "Apache-2.0",
"bin": {
"openai": "bin/cli"
@@ -3994,9 +3995,9 @@
"license": "MIT"
},
"node_modules/semver": {
"version": "7.7.3",
"resolved": "https://registry.npmjs.org/semver/-/semver-7.7.3.tgz",
"integrity": "sha512-SdsKMrI9TdgjdweUSR9MweHA4EJ8YxHn8DFaDisvhVlUOe4BF1tLD7GAj0lIqWVl+dPb/rExr0Btby5loQm20Q==",
"version": "7.7.4",
"resolved": "https://registry.npmjs.org/semver/-/semver-7.7.4.tgz",
"integrity": "sha512-vFKC2IEtQnVhpT78h1Yp8wzwrf8CM+MzKMHGJZfBtzhZNycRFnXsHk6E5TxIkkMsgNS7mdX3AGB7x2QM2di4lA==",
"license": "ISC",
"bin": {
"semver": "bin/semver.js"
@@ -4403,9 +4404,9 @@
"license": "MIT"
},
"node_modules/undici": {
"version": "7.19.2",
"resolved": "https://registry.npmjs.org/undici/-/undici-7.19.2.tgz",
"integrity": "sha512-4VQSpGEGsWzk0VYxyB/wVX/Q7qf9t5znLRgs0dzszr9w9Fej/8RVNQ+S20vdXSAyra/bJ7ZQfGv6ZMj7UEbzSg==",
"version": "7.21.0",
"resolved": "https://registry.npmjs.org/undici/-/undici-7.21.0.tgz",
"integrity": "sha512-Hn2tCQpoDt1wv23a68Ctc8Cr/BHpUSfaPYrkajTXOS9IKpxVRx/X5m1K2YkbK2ipgZgxXSgsUinl3x+2YdSSfg==",
"license": "MIT",
"engines": {
"node": ">=20.18.1"
@@ -4667,6 +4668,18 @@
"integrity": "sha512-zksaLKM2fVlnB5jQQDqKXXwYHLQUVH9es+5TOOHwGOVJOCeRBCiPjwSg+3tN2AdTCzjgli4jijCH290kXb/zWQ==",
"license": "Apache-2.0"
},
"node_modules/wavefile": {
"version": "11.0.0",
"resolved": "https://registry.npmjs.org/wavefile/-/wavefile-11.0.0.tgz",
"integrity": "sha512-/OBiAALgWU24IG7sC84cDO/KfFuvajWc5Uec0oV2zrpOOZZDgGdOwHwgEzOrwh8jkubBk7PtZfQBIcI1OaE5Ng==",
"license": "MIT",
"bin": {
"wavefile": "bin/wavefile.js"
},
"engines": {
"node": ">=8"
}
},
"node_modules/webidl-conversions": {
"version": "3.0.1",
"resolved": "https://registry.npmjs.org/webidl-conversions/-/webidl-conversions-3.0.1.tgz",

View File

@@ -1,6 +1,6 @@
{
"name": "@ztimson/ai-utils",
"version": "0.6.1",
"version": "0.7.3",
"description": "AI Utility library",
"author": "Zak Timson",
"license": "MIT",
@@ -32,7 +32,8 @@
"@ztimson/utils": "^0.27.9",
"cheerio": "^1.2.0",
"openai": "^6.6.0",
"tesseract.js": "^6.0.1"
"tesseract.js": "^6.0.1",
"wavefile": "^11.0.0"
},
"devDependencies": {
"@types/node": "^24.8.1",

View File

@@ -8,6 +8,8 @@ export type AbortablePromise<T> = Promise<T> & {
};
export type AiOptions = {
/** Token to pull models from hugging face */
hfToken?: string;
/** Path to models */
path?: string;
/** ASR model: whisper-tiny, whisper-base */

134
src/asr.ts Normal file
View File

@@ -0,0 +1,134 @@
import { pipeline } from '@xenova/transformers';
import { parentPort } from 'worker_threads';
import { spawn } from 'node:child_process';
import { execSync } from 'node:child_process';
import { mkdtempSync, rmSync, readFileSync } from 'node:fs';
import { join } from 'node:path';
import { tmpdir } from 'node:os';
import wavefile from 'wavefile';
let whisperPipeline: any;
export async function canDiarization(): Promise<boolean> {
return new Promise((resolve) => {
const proc = spawn('python', ['-c', 'import pyannote.audio']);
proc.on('close', (code: number) => resolve(code === 0));
proc.on('error', () => resolve(false));
});
}
async function runDiarization(audioPath: string, dir: string, token: string): Promise<any[]> {
const script = `
import sys
import json
import os
from pyannote.audio import Pipeline
os.environ['TORCH_HOME'] = r"${dir}"
pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization-3.1", token="${token}")
output = pipeline(sys.argv[1])
segments = []
for turn, speaker in output.speaker_diarization:
segments.append({"start": turn.start, "end": turn.end, "speaker": speaker})
print(json.dumps(segments))
`;
return new Promise((resolve, reject) => {
let output = '';
const proc = spawn('python', ['-c', script, audioPath]);
proc.stdout.on('data', (data: Buffer) => output += data.toString());
proc.stderr.on('data', (data: Buffer) => console.error(data.toString()));
proc.on('close', (code: number) => {
if(code === 0) {
try {
resolve(JSON.parse(output));
} catch (err) {
reject(new Error('Failed to parse diarization output'));
}
} else {
reject(new Error(`Python process exited with code ${code}`));
}
});
proc.on('error', reject);
});
}
function combineSpeakerTranscript(chunks: any[], speakers: any[]): string {
const speakerMap = new Map();
let speakerCount = 0;
speakers.forEach((seg: any) => {
if(!speakerMap.has(seg.speaker)) speakerMap.set(seg.speaker, ++speakerCount);
});
const lines: string[] = [];
let currentSpeaker = -1;
let currentText = '';
chunks.forEach((chunk: any) => {
const time = chunk.timestamp[0];
const speaker = speakers.find((s: any) => time >= s.start && time <= s.end);
const speakerNum = speaker ? speakerMap.get(speaker.speaker) : 1;
if (speakerNum !== currentSpeaker) {
if(currentText) lines.push(`[Speaker ${currentSpeaker}]: ${currentText.trim()}`);
currentSpeaker = speakerNum;
currentText = chunk.text;
} else {
currentText += chunk.text;
}
});
if(currentText) lines.push(`[Speaker ${currentSpeaker}]: ${currentText.trim()}`);
return lines.join('\n');
}
function prepareAudioBuffer(file: string): [string, Float32Array] {
let wav: any, tmp;
try {
wav = new wavefile.WaveFile(readFileSync(file));
} catch(err) {
tmp = join(mkdtempSync(join(tmpdir(), 'audio-')), 'converted.wav');
execSync(`ffmpeg -i "${file}" -ar 16000 -ac 1 -f wav "${tmp}"`, { stdio: 'ignore' });
wav = new wavefile.WaveFile(readFileSync(tmp));
} finally {
wav.toBitDepth('32f');
wav.toSampleRate(16000);
const samples = wav.getSamples();
if(Array.isArray(samples)) {
const left = samples[0];
const right = samples[1];
const buffer = new Float32Array(left.length);
for (let i = 0; i < left.length; i++) buffer[i] = (left[i] + right[i]) / 2;
return [tmp || file, buffer];
}
return [tmp || file, samples];
}
}
parentPort?.on('message', async ({ file, speaker, model, modelDir, token }) => {
try {
if(!whisperPipeline) whisperPipeline = await pipeline('automatic-speech-recognition', `Xenova/${model}`, {cache_dir: modelDir, quantized: true});
// Prepare audio file
const [f, buffer] = prepareAudioBuffer(file);
// Fetch transcript and speakers
const hasDiarization = speaker && await canDiarization();
const [transcript, speakers] = await Promise.all([
whisperPipeline(buffer, {return_timestamps: speaker ? 'word' : false}),
(!speaker || !token || !hasDiarization) ? Promise.resolve(): runDiarization(f, modelDir, token),
]);
if(file != f) rmSync(f, { recursive: true, force: true });
// Return any results / errors if no more processing required
const text = transcript.text?.trim() || null;
if(!speaker) return parentPort?.postMessage({ text });
if(!token) return parentPort?.postMessage({ text, error: 'HuggingFace token required' });
if(!hasDiarization) return parentPort?.postMessage({ text, error: 'Speaker diarization unavailable' });
// Combine transcript and speakers
const combined = combineSpeakerTranscript(transcript.chunks || [], speakers || []);
parentPort?.postMessage({ text: combined });
} catch (err: any) {
parentPort?.postMessage({ error: err.stack || err.message });
}
});

View File

@@ -1,115 +1,60 @@
import { spawn } from 'node:child_process';
import { pipeline } from '@xenova/transformers';
import {fileURLToPath} from 'url';
import {Worker} from 'worker_threads';
import {AbortablePromise, Ai} from './ai.ts';
import {canDiarization} from './asr.ts';
import {dirname, join} from 'path';
export class Audio {
private whisperPipeline: any;
constructor(private ai: Ai) {}
private combineSpeakerTranscript(chunks: any[], speakers: any[]): string {
const speakerMap = new Map();
let speakerCount = 0;
speakers.forEach((seg: any) => {
if(!speakerMap.has(seg.speaker)) speakerMap.set(seg.speaker, ++speakerCount);
});
const lines: string[] = [];
let currentSpeaker = -1;
let currentText = '';
chunks.forEach((chunk: any) => {
const time = chunk.timestamp[0];
const speaker = speakers.find((s: any) => time >= s.start && time <= s.end);
const speakerNum = speaker ? speakerMap.get(speaker.speaker) : 1;
if (speakerNum !== currentSpeaker) {
if(currentText) lines.push(`[speaker ${currentSpeaker}]: ${currentText.trim()}`);
currentSpeaker = speakerNum;
currentText = chunk.text;
} else {
currentText += chunk.text;
}
});
if(currentText) lines.push(`[speaker ${currentSpeaker}]: ${currentText.trim()}`);
return lines.join('\n');
}
async canDiarization(): Promise<boolean> {
return new Promise((resolve) => {
const proc = spawn('python3', ['-c', 'import pyannote.audio']);
proc.on('close', (code: number) => resolve(code === 0));
proc.on('error', () => resolve(false));
});
}
private async runDiarization(audioPath: string): Promise<any[]> {
if(!await this.canDiarization()) throw new Error('Pyannote is not installed: pip install pyannote.audio');
const script = `
import sys
import json
from pyannote.audio import Pipeline
os.environ['TORCH_HOME'] = "${this.ai.options.path}"
pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization-3.1")
diarization = pipeline(sys.argv[1])
segments = []
for turn, _, speaker in diarization.itertracks(yield_label=True):
segments.append({
"start": turn.start,
"end": turn.end,
"speaker": speaker
})
print(json.dumps(segments))
`;
return new Promise((resolve, reject) => {
let output = '';
const proc = spawn('python3', ['-c', script, audioPath]);
proc.stdout.on('data', (data: Buffer) => output += data.toString());
proc.stderr.on('data', (data: Buffer) => console.error(data.toString()));
proc.on('close', (code: number) => {
if(code === 0) {
try {
resolve(JSON.parse(output));
} catch (err) {
reject(new Error('Failed to parse diarization output'));
}
} else {
reject(new Error(`Python process exited with code ${code}`));
}
});
proc.on('error', reject);
});
}
asr(path: string, options: { model?: string; speaker?: boolean } = {}): AbortablePromise<string | null> {
asr(file: string, options: { model?: string; speaker?: boolean | 'id' } = {}): AbortablePromise<string | null> {
const { model = this.ai.options.asr || 'whisper-base', speaker = false } = options;
let aborted = false;
const abort = () => { aborted = true; };
const p = new Promise<string | null>(async (resolve, reject) => {
try {
if(aborted) return resolve(null);
if(!this.whisperPipeline) this.whisperPipeline = await pipeline('automatic-speech-recognition', `Xenova/${model}`, { cache_dir: this.ai.options.path, quantized: true });
// Transcript
if(aborted) return resolve(null);
const transcriptResult = await this.whisperPipeline(path, {return_timestamps: speaker ? 'word' : false, chunk_length_s: 30,});
if(!speaker) return resolve(transcriptResult.text?.trim() || null);
// Speaker Diarization
if(aborted) return resolve(null);
const speakers = await this.runDiarization(path);
if(aborted) return resolve(null);
const combined = this.combineSpeakerTranscript(transcriptResult.chunks || [], speakers);
resolve(combined);
} catch (err) {
reject(err);
let p = new Promise<string | null>((resolve, reject) => {
const worker = new Worker(join(dirname(fileURLToPath(import.meta.url)), 'asr.js'));
const handleMessage = ({ text, warning, error }: any) => {
worker.terminate();
if(aborted) return;
if(error) reject(new Error(error));
else {
if(warning) console.warn(warning);
resolve(text);
}
};
const handleError = (err: Error) => {
worker.terminate();
if(!aborted) reject(err);
};
worker.on('message', handleMessage);
worker.on('error', handleError);
worker.on('exit', (code) => {
if(code !== 0 && !aborted) reject(new Error(`Worker exited with code ${code}`));
});
worker.postMessage({file, model, speaker, modelDir: this.ai.options.path, token: this.ai.options.hfToken});
});
// Name speakers using AI
if(options.speaker == 'id') {
if(!this.ai.language.defaultModel) throw new Error('Configure an LLM for advanced ASR speaker detection');
p = p.then(async transcript => {
if(!transcript) return transcript;
let chunks = this.ai.language.chunk(transcript, 500, 0);
if(chunks.length > 4) chunks = [...chunks.slice(0, 3), <string>chunks.at(-1)];
const names = await this.ai.language.json(chunks.join('\n'), '{1: "Detected Name"}', {
system: 'Use this following transcript to identify speakers. Only identify speakers you are sure about',
temperature: 0.1,
});
Object.entries(names).forEach(([speaker, name]) => {
transcript = (<string>transcript).replaceAll(`[Speaker ${speaker}]`, `[${name}]`);
});
return transcript;
})
}
return Object.assign(p, { abort });
}
canDiarization = canDiarization;
}

View File

@@ -3,12 +3,9 @@ import { parentPort } from 'worker_threads';
let embedder: any;
parentPort?.on('message', async ({ id, text, model, path }) => {
if(!embedder) embedder = await pipeline('feature-extraction', 'Xenova/' + model, {
quantized: true,
cache_dir: path,
});
parentPort?.on('message', async ({text, model, modelDir }) => {
if(!embedder) embedder = await pipeline('feature-extraction', 'Xenova/' + model, {quantized: true, cache_dir: modelDir});
const output = await embedder(text, { pooling: 'mean', normalize: true });
const embedding = Array.from(output.data);
parentPort?.postMessage({ id, embedding });
parentPort?.postMessage({embedding});
});

View File

@@ -1,5 +1,6 @@
export * from './ai';
export * from './antrhopic';
export * from './asr';
export * from './audio';
export * from './embedder'
export * from './llm';

View File

@@ -75,22 +75,10 @@ export type LLMRequest = {
}
class LLM {
private embedWorker: Worker | null = null;
private embedQueue = new Map<number, { resolve: (value: number[]) => void; reject: (error: any) => void }>();
private embedId = 0;
private models: {[model: string]: LLMProvider} = {};
private defaultModel!: string;
defaultModel!: string;
models: {[model: string]: LLMProvider} = {};
constructor(public readonly ai: Ai) {
this.embedWorker = new Worker(join(dirname(fileURLToPath(import.meta.url)), 'embedder.js'));
this.embedWorker.on('message', ({ id, embedding }) => {
const pending = this.embedQueue.get(id);
if (pending) {
pending.resolve(embedding);
this.embedQueue.delete(id);
}
});
if(!ai.options.llm?.models) return;
Object.entries(ai.options.llm.models).forEach(([model, config]) => {
if(!this.defaultModel) this.defaultModel = model;
@@ -196,7 +184,12 @@ class LLM {
const system = history[0].role == 'system' ? history[0] : null,
recent = keep == 0 ? [] : history.slice(-keep),
process = (keep == 0 ? history : history.slice(0, -keep)).filter(h => h.role === 'assistant' || h.role === 'user');
const summary: any = await this.json(`Create the smallest summary possible, no more than 500 tokens. Create a list of NEW facts (split by subject [pro]noun and fact) about what you learned from this conversation that you didn't already know or get from a tool call or system prompt. Focus only on new information about people, topics, or facts. Avoid generating facts about the AI. Match this format: {summary: string, facts: [[subject, fact]]}\n\n${process.map(m => `${m.role}: ${m.content}`).join('\n\n')}`, {model: options?.model, temperature: options?.temperature || 0.3});
const summary: any = await this.json(process.map(m => `${m.role}: ${m.content}`).join('\n\n'), '{summary: string, facts: [[subject, fact]]}', {
system: 'Create the smallest summary possible, no more than 500 tokens. Create a list of NEW facts (split by subject [pro]noun and fact) about what you learned from this conversation that you didn\'t already know or get from a tool call or system prompt. Focus only on new information about people, topics, or facts. Avoid generating facts about the AI.',
model: options?.model,
temperature: options?.temperature || 0.3
});
const timestamp = new Date();
const memory = await Promise.all((summary?.facts || [])?.map(async ([owner, fact]: [string, string]) => {
const e = await Promise.all([this.embedding(owner), this.embedding(`${owner}: ${fact}`)]);
@@ -262,30 +255,37 @@ class LLM {
/**
* Create a vector representation of a string
* @param {object | string} target Item that will be embedded (objects get converted)
* @param {number} maxTokens Chunking size. More = better context, less = more specific (Search by paragraphs or lines)
* @param {number} overlapTokens Includes previous X tokens to provide continuity to AI (In addition to max tokens)
* @param {maxTokens?: number, overlapTokens?: number} opts Options for embedding such as chunk sizes
* @returns {Promise<Awaited<{index: number, embedding: number[], text: string, tokens: number}>[]>} Chunked embeddings
*/
embedding(target: object | string, maxTokens = 500, overlapTokens = 50) {
async embedding(target: object | string, opts: {maxTokens?: number, overlapTokens?: number} = {}) {
let {maxTokens = 500, overlapTokens = 50} = opts;
const embed = (text: string): Promise<number[]> => {
return new Promise((resolve, reject) => {
const id = this.embedId++;
this.embedQueue.set(id, { resolve, reject });
this.embedWorker?.postMessage({
id,
text,
model: this.ai.options?.embedder || 'bge-small-en-v1.5',
path: this.ai.options.path
const worker = new Worker(join(dirname(fileURLToPath(import.meta.url)), 'embedder.js'));
const handleMessage = ({ embedding }: any) => {
worker.terminate();
resolve(embedding);
};
const handleError = (err: Error) => {
worker.terminate();
reject(err);
};
worker.on('message', handleMessage);
worker.on('error', handleError);
worker.on('exit', (code) => {
if(code !== 0) reject(new Error(`Worker exited with code ${code}`));
});
worker.postMessage({text, model: this.ai.options?.embedder || 'bge-small-en-v1.5', modelDir: this.ai.options.path});
});
};
const chunks = this.chunk(target, maxTokens, overlapTokens);
return Promise.all(chunks.map(async (text, index) => ({
index,
embedding: await embed(text),
text,
tokens: this.estimateTokens(text),
})));
const chunks = this.chunk(target, maxTokens, overlapTokens), results: any[] = [];
for(let i = 0; i < chunks.length; i++) {
const text= chunks[i];
const embedding = await embed(text);
results.push({index: i, embedding, text, tokens: this.estimateTokens(text)});
}
return results;
}
/**
@@ -317,12 +317,16 @@ class LLM {
/**
* Ask a question with JSON response
* @param {string} message Question
* @param {string} text Text to process
* @param {string} schema JSON schema the AI should match
* @param {LLMRequest} options Configuration options and chat history
* @returns {Promise<{} | {} | RegExpExecArray | null>}
*/
async json(message: string, options?: LLMRequest): Promise<any> {
let resp = await this.ask(message, {system: 'Respond using a JSON blob matching any provided examples', ...options});
async json(text: string, schema: string, options?: LLMRequest): Promise<any> {
let resp = await this.ask(text, {...options, system: (options?.system ? `${options.system}\n` : '') + `Only respond using a JSON code block matching this schema:
\`\`\`json
${schema}
\`\`\``});
if(!resp) return {};
const codeBlock = /```(?:.+)?\s*([\s\S]*?)```/.exec(resp);
const jsonStr = codeBlock ? codeBlock[1].trim() : resp;

View File

@@ -6,6 +6,7 @@ export default defineConfig({
build: {
lib: {
entry: {
asr: './src/asr.ts',
index: './src/index.ts',
embedder: './src/embedder.ts',
},