Archived

Bumps [pyright](https://github.com/Microsoft/pyright/tree/HEAD/packages/pyright) from 1.1.296 to 1.1.301.
- [Release notes](https://github.com/Microsoft/pyright/releases)
- [Commits](https://github.com/Microsoft/pyright/commits/1.1.301/packages/pyright)

---
updated-dependencies:
- dependency-name: pyright
  dependency-type: direct:development
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>

.github/dependabot.yml

@@ -4,12 +4,12 @@ updates:
   - package-ecosystem: "npm"
     directory: "/"
     schedule:
-      interval: "weekly"
+      interval: "monthly"
   - package-ecosystem: "pip"
     directory: "/"
     schedule:
-      interval: "weekly"
+      interval: "monthly"
   - package-ecosystem: "github-actions"
     directory: "/"
     schedule:
-      interval: "weekly"
+      interval: "monthly"

.github/workflows/ci.yml

@@ -18,9 +18,9 @@ jobs:
     steps:
       - uses: actions/checkout@v3
       - name: Get tags
-        run: git fetch --tags origin
+        run: git fetch --tags origin -f
 
-      - uses: awalsh128/cache-apt-pkgs-action@v1.1.1
+      - uses: awalsh128/cache-apt-pkgs-action@v1.3.0
         if: runner.os == 'Linux'
         with:
           packages: portaudio19-dev

.github/workflows/stale-issues.yml

@@ -7,7 +7,7 @@ jobs:
   close-issues:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/stale@v6.0.1
+      - uses: actions/stale@v8.0.0
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
           stale-issue-message: "This issue is stale because it has been open for 21 days with no activity."

.github/workflows/typos.yml

@@ -18,4 +18,4 @@ jobs:
       - uses: actions/checkout@v3
 
       - name: typos-action
-        uses: crate-ci/typos@v1.12.8
+        uses: crate-ci/typos@v1.14.3

.gitignore

@@ -162,3 +162,5 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
+
+.vscode/

Makefile

@@ -35,3 +35,7 @@ markdownlint:
 
 lint_node: markdownlint pyright
 
+
+style:
+	find $(TARGET_DIRS) | grep '\.py$$' | xargs black
+	find $(TARGET_DIRS) | grep '\.py$$' | xargs isort

README.md

@@ -1,7 +1,7 @@
 
 # Whispering
 
-[![MIT License](https://img.shields.io/apm/l/atomic-design-ui.svg?)](LICENSE)
+[![MIT License](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
 ![Python Versions](https://img.shields.io/badge/python-3.8%20%7C%203.9%20%7C%203.10-blue)
 
 [![CI](https://github.com/shirayu/whispering/actions/workflows/ci.yml/badge.svg)](https://github.com/shirayu/whispering/actions/workflows/ci.yml)
@@ -11,10 +11,18 @@
 Streaming transcriber with [whisper](https://github.com/openai/whisper).
 Enough machine power is needed to transcribe in real time.
 
+## Notice
+
+This repository has been archived.
+There are some alternatives.
+
+- <https://github.com/guillaumekln/faster-whisper>
+- <https://github.com/ggerganov/whisper.cpp>
+
 ## Setup
 
 ```bash
-pip install -U git+https://github.com/shirayu/whispering.git@v0.5.1
+pip install -U git+https://github.com/shirayu/whispering.git@v0.6.6
 
 # If you use GPU, install proper torch and torchaudio
 # Check https://pytorch.org/get-started/locally/
@@ -32,30 +40,31 @@ sudo apt -y install portaudio19-dev
 
 ```bash
 # Run in English
+#  By the default, it needs to wait at least 30 seconds
 whispering --language en --model tiny
 ```
 
 - ``--help`` shows full options
-- ``--model`` set the [model name](https://github.com/openai/whisper#available-models-and-languages) to use. Larger models will be more accurate, but may not be able to transcribe in real time.
+- ``--model`` sets the [model name](https://github.com/openai/whisper#available-models-and-languages) to use. Larger models will be more accurate, but may not be able to transcribe in real time.
 - ``--language`` sets the language to transcribe. The list of languages are shown with ``whispering -h``
 - ``--no-progress`` disables the progress message
 - ``-t`` sets temperatures to decode. You can set several like ``-t 0.0 -t 0.1 -t 0.5``, but too many temperatures exhaust decoding time
 - ``--debug`` outputs logs for debug
-- ``--no-vad`` disables VAD (Voice Activity Detection). This forces whisper to analyze non-voice activity sound period
+- ``--vad`` sets VAD (Voice Activity Detection) threshold. The default is ``0.5``. ``0`` disables VAD and forces whisper to analyze non-voice activity sound period. Try ``--vad 0`` if VAD prevents transcription.
 - ``--output`` sets output file (Default: Standard output)
+- ``--frame``: the number of minimum frames of mel spectrogram input for Whisper (default: ``3000``. i.e. 30 seconds)
 
 ### Parse interval
 
-Without ``--allow-padding``, whispering just performs VAD for the period,
-and when it is predicted as "silence", it will not be passed to whisper.
-If you want to change the VAD interval, change ``-n``.
+By default, whispering performs VAD for every 3.75 second.
+This interval is determined by the value of ``-n`` and its default is ``20``.
+When an interval is predicted as "silence", it will not be passed to whisper.
+If you want to disable VAD, please make VAD threshold 0 by adding ``--vad 0``.
 
-If you want quick response, set small ``-n`` and add ``--allow-padding``.
-However, this may sacrifice the accuracy.
-
-```bash
-whispering --language en --model tiny -n 20 --allow-padding
-```
+By default, whispering does not perform analysis until the total length of the segments determined by VAD to have speech exceeds 30 seconds.
+This is because the original Whisper assumes that the inputs are 30 seconds segments.
+However, if silence segments appear 16 times (the default value of ``--max_nospeech_skip``) after speech is detected, the analysis is performed.
+You can make the length of segments smaller with ``--frame`` option (default: 3000), but it sacrifices accuracy because this is not expected input for Whisper.
 
 ## Example of web socket
 
@@ -75,7 +84,53 @@ whispering --language en --model tiny --host 0.0.0.0 --port 8000
 whispering --host ADDRESS_OF_HOST --port 8000 --mode client
 ```
 
-You can set ``-n``, ``--allow-padding`` and other options.
+You can set ``-n`` and other options.
+
+## For Developers
+
+1. Install [Python](https://www.python.org/) and [Node.js](https://nodejs.org/)
+2. [Install poetry](https://python-poetry.org/docs/) to use ``poetry`` command
+3. Clone and install libraries
+
+    ```console
+    # Clone
+    git clone https://github.com/shirayu/whispering.git
+
+    # With poetry
+    poetry config virtualenvs.in-project true
+    poetry install --all-extras
+    poetry run pip install -U torch torchaudio --extra-index-url https://download.pytorch.org/whl/cu116
+
+    # With npm
+    npm install
+    ```
+
+4. Run test and check that no errors occur
+
+    ```bash
+    poetry run make -j4
+    ```
+
+5. Make fancy updates
+6. Make style
+
+    ```bash
+    poetry run make style
+    ```
+
+7. Run test again and check that no errors occur
+
+    ```bash
+    poetry run make -j4
+    ```
+
+8. Check typos by using [typos](https://github.com/crate-ci/typos). Just run ``typos`` command in the root directory.
+
+    ```bash
+    typos
+    ```
+
+9. Send Pull requests!
 
 ## License
 

package-lock.json

@@ -1,15 +1,15 @@
 {
     "name": "pyright-exec",
     "version": "1.0.0",
-    "lockfileVersion": 2,
+    "lockfileVersion": 3,
     "requires": true,
     "packages": {
         "": {
             "name": "pyright-exec",
             "version": "1.0.0",
             "devDependencies": {
-                "markdownlint-cli": "^0.32.1",
-                "pyright": "^1.1.274"
+                "markdownlint-cli": "^0.33.0",
+                "pyright": "^1.1.301"
             }
         },
         "node_modules/argparse": {
@@ -34,9 +34,9 @@
             }
         },
         "node_modules/commander": {
-            "version": "9.4.0",
-            "resolved": "https://registry.npmjs.org/commander/-/commander-9.4.0.tgz",
-            "integrity": "sha512-sRPT+umqkz90UA8M1yqYfnHlZA7fF6nSphDtxeywPZ49ysjxDQybzk13CL+mXekDRG92skbcqCLVovuCusNmFw==",
+            "version": "9.4.1",
+            "resolved": "https://registry.npmjs.org/commander/-/commander-9.4.1.tgz",
+            "integrity": "sha512-5EEkTNyHNGFPD2H+c/dXXfQZYa/scCKasxWcXJaWnNJ99pnQN9Vnmqow+p+PlFPE63Q6mThaZws1T+HxfpgtPw==",
             "dev": true,
             "engines": {
                 "node": "^12.20.0 || >=14"
@@ -101,9 +101,9 @@
             }
         },
         "node_modules/ignore": {
-            "version": "5.2.0",
-            "resolved": "https://registry.npmjs.org/ignore/-/ignore-5.2.0.tgz",
-            "integrity": "sha512-CmxgYGiEPCLhfLnpPp1MoRmifwEIOgjcHXxOBjv7mY96c+eWScsOP9c112ZyLdWHi0FxHjI+4uVhKYp/gcdRmQ==",
+            "version": "5.2.4",
+            "resolved": "https://registry.npmjs.org/ignore/-/ignore-5.2.4.tgz",
+            "integrity": "sha512-MAb38BcSbH0eHNBxn7ql2NH/kX33OkB3lZ1BNdh7ENeRChHTYsTvWrMubiIAMNS2llXEEgZ1MUOBtXChP3kaFQ==",
             "dev": true,
             "engines": {
                 "node": ">= 4"
@@ -147,9 +147,9 @@
             }
         },
         "node_modules/jsonc-parser": {
-            "version": "3.1.0",
-            "resolved": "https://registry.npmjs.org/jsonc-parser/-/jsonc-parser-3.1.0.tgz",
-            "integrity": "sha512-DRf0QjnNeCUds3xTjKlQQ3DpJD51GvDjJfnxUVWg6PZTo2otSm+slzNAxU/35hF8/oJIKoG9slq30JYOsF2azg==",
+            "version": "3.2.0",
+            "resolved": "https://registry.npmjs.org/jsonc-parser/-/jsonc-parser-3.2.0.tgz",
+            "integrity": "sha512-gfFQZrcTc8CnKXp6Y4/CBT3fTc0OVuDofpre4aEeEpSBPV5X5v4+Vmx+8snU7RLPrNHPKSgLxGo9YuQzz20o+w==",
             "dev": true
         },
         "node_modules/linkify-it": {
@@ -178,32 +178,31 @@
             }
         },
         "node_modules/markdownlint": {
-            "version": "0.26.2",
-            "resolved": "https://registry.npmjs.org/markdownlint/-/markdownlint-0.26.2.tgz",
-            "integrity": "sha512-2Am42YX2Ex5SQhRq35HxYWDfz1NLEOZWWN25nqd2h3AHRKsGRE+Qg1gt1++exW792eXTrR4jCNHfShfWk9Nz8w==",
+            "version": "0.27.0",
+            "resolved": "https://registry.npmjs.org/markdownlint/-/markdownlint-0.27.0.tgz",
+            "integrity": "sha512-HtfVr/hzJJmE0C198F99JLaeada+646B5SaG2pVoEakLFI6iRGsvMqrnnrflq8hm1zQgwskEgqSnhDW11JBp0w==",
             "dev": true,
             "dependencies": {
                 "markdown-it": "13.0.1"
             },
             "engines": {
-                "node": ">=14"
+                "node": ">=14.18.0"
             }
         },
         "node_modules/markdownlint-cli": {
-            "version": "0.32.2",
-            "resolved": "https://registry.npmjs.org/markdownlint-cli/-/markdownlint-cli-0.32.2.tgz",
-            "integrity": "sha512-xmJT1rGueUgT4yGNwk6D0oqQr90UJ7nMyakXtqjgswAkEhYYqjHew9RY8wDbOmh2R270IWjuKSeZzHDEGPAUkQ==",
+            "version": "0.33.0",
+            "resolved": "https://registry.npmjs.org/markdownlint-cli/-/markdownlint-cli-0.33.0.tgz",
+            "integrity": "sha512-zMK1oHpjYkhjO+94+ngARiBBrRDEUMzooDHBAHtmEIJ9oYddd9l3chCReY2mPlecwH7gflQp1ApilTo+o0zopQ==",
             "dev": true,
             "dependencies": {
-                "commander": "~9.4.0",
+                "commander": "~9.4.1",
                 "get-stdin": "~9.0.0",
                 "glob": "~8.0.3",
-                "ignore": "~5.2.0",
+                "ignore": "~5.2.4",
                 "js-yaml": "^4.1.0",
-                "jsonc-parser": "~3.1.0",
-                "markdownlint": "~0.26.2",
-                "markdownlint-rule-helpers": "~0.17.2",
-                "minimatch": "~5.1.0",
+                "jsonc-parser": "~3.2.0",
+                "markdownlint": "~0.27.0",
+                "minimatch": "~5.1.2",
                 "run-con": "~1.2.11"
             },
             "bin": {
@@ -213,15 +212,6 @@
                 "node": ">=14"
             }
         },
-        "node_modules/markdownlint-rule-helpers": {
-            "version": "0.17.2",
-            "resolved": "https://registry.npmjs.org/markdownlint-rule-helpers/-/markdownlint-rule-helpers-0.17.2.tgz",
-            "integrity": "sha512-XaeoW2NYSlWxMCZM2B3H7YTG6nlaLfkEZWMBhr4hSPlq9MuY2sy83+Xr89jXOqZMZYjvi5nBCGoFh7hHoPKZmA==",
-            "dev": true,
-            "engines": {
-                "node": ">=12"
-            }
-        },
         "node_modules/mdurl": {
             "version": "1.0.1",
             "resolved": "https://registry.npmjs.org/mdurl/-/mdurl-1.0.1.tgz",
@@ -229,9 +219,9 @@
             "dev": true
         },
         "node_modules/minimatch": {
-            "version": "5.1.0",
-            "resolved": "https://registry.npmjs.org/minimatch/-/minimatch-5.1.0.tgz",
-            "integrity": "sha512-9TPBGGak4nHfGZsPBohm9AWg6NoT7QTCehS3BIJABslyZbzxfV78QM2Y6+i741OPZIafFAaiiEMh5OyIrJPgtg==",
+            "version": "5.1.6",
+            "resolved": "https://registry.npmjs.org/minimatch/-/minimatch-5.1.6.tgz",
+            "integrity": "sha512-lKwV/1brpG6mBUFHtb7NUmtABCb2WZZmm2wNiOA5hAb8VdCS4B3dtMWyvcoViccwAW/COERjXLt0zP1zXUN26g==",
             "dev": true,
             "dependencies": {
                 "brace-expansion": "^2.0.1"
@@ -241,10 +231,13 @@
             }
         },
         "node_modules/minimist": {
-            "version": "1.2.6",
-            "resolved": "https://registry.npmjs.org/minimist/-/minimist-1.2.6.tgz",
-            "integrity": "sha512-Jsjnk4bw3YJqYzbdyBiNsPWHPfO++UGG749Cxs6peCu5Xg4nrena6OVxOYxrQTqww0Jmwt+Ref8rggumkTLz9Q==",
-            "dev": true
+            "version": "1.2.8",
+            "resolved": "https://registry.npmjs.org/minimist/-/minimist-1.2.8.tgz",
+            "integrity": "sha512-2yyAR8qBkN3YuheJanUpWC5U3bb5osDywNB8RzDVlDwDHbocAJveqqj1u8+SVD7jkWT4yvsHCpWqqWqAxb0zCA==",
+            "dev": true,
+            "funding": {
+                "url": "https://github.com/sponsors/ljharb"
+            }
         },
         "node_modules/once": {
             "version": "1.4.0",
@@ -256,9 +249,9 @@
             }
         },
         "node_modules/pyright": {
-            "version": "1.1.274",
-            "resolved": "https://registry.npmjs.org/pyright/-/pyright-1.1.274.tgz",
-            "integrity": "sha512-+MEiHktoAxlehWbBAF2vwJkJxV5tObbLK6vnc7u+iDBH06vuHqTqY33AV+YGNwfDEU35chw5zu7H3EpGROXsGQ==",
+            "version": "1.1.301",
+            "resolved": "https://registry.npmjs.org/pyright/-/pyright-1.1.301.tgz",
+            "integrity": "sha512-Y4MMELxQ/5+/FlWjbQTg5wbP3z+V4IyFcATSsNLpZbJm0y4gz6ijf/b0zZV1sA8yJstf6xJ98vw5qxPM0yU8Zg==",
             "dev": true,
             "bin": {
                 "pyright": "index.js",
@@ -307,235 +300,5 @@
             "integrity": "sha512-l4Sp/DRseor9wL6EvV2+TuQn63dMkPjZ/sp9XkghTEbV9KlPS1xUsZ3u7/IQO4wxtcFB4bgpQPRcR3QCvezPcQ==",
             "dev": true
         }
-    },
-    "dependencies": {
-        "argparse": {
-            "version": "2.0.1",
-            "resolved": "https://registry.npmjs.org/argparse/-/argparse-2.0.1.tgz",
-            "integrity": "sha512-8+9WqebbFzpX9OR+Wa6O29asIogeRMzcGtAINdpMHHyAg10f05aSFVBbcEqGf/PXw1EjAZ+q2/bEBg3DvurK3Q==",
-            "dev": true
-        },
-        "balanced-match": {
-            "version": "1.0.2",
-            "resolved": "https://registry.npmjs.org/balanced-match/-/balanced-match-1.0.2.tgz",
-            "integrity": "sha512-3oSeUO0TMV67hN1AmbXsK4yaqU7tjiHlbxRDZOpH0KW9+CeX4bRAaX0Anxt0tx2MrpRpWwQaPwIlISEJhYU5Pw==",
-            "dev": true
-        },
-        "brace-expansion": {
-            "version": "2.0.1",
-            "resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-2.0.1.tgz",
-            "integrity": "sha512-XnAIvQ8eM+kC6aULx6wuQiwVsnzsi9d3WxzV3FpWTGA19F621kwdbsAcFKXgKUHZWsy+mY6iL1sHTxWEFCytDA==",
-            "dev": true,
-            "requires": {
-                "balanced-match": "^1.0.0"
-            }
-        },
-        "commander": {
-            "version": "9.4.0",
-            "resolved": "https://registry.npmjs.org/commander/-/commander-9.4.0.tgz",
-            "integrity": "sha512-sRPT+umqkz90UA8M1yqYfnHlZA7fF6nSphDtxeywPZ49ysjxDQybzk13CL+mXekDRG92skbcqCLVovuCusNmFw==",
-            "dev": true
-        },
-        "deep-extend": {
-            "version": "0.6.0",
-            "resolved": "https://registry.npmjs.org/deep-extend/-/deep-extend-0.6.0.tgz",
-            "integrity": "sha512-LOHxIOaPYdHlJRtCQfDIVZtfw/ufM8+rVj649RIHzcm/vGwQRXFt6OPqIFWsm2XEMrNIEtWR64sY1LEKD2vAOA==",
-            "dev": true
-        },
-        "entities": {
-            "version": "3.0.1",
-            "resolved": "https://registry.npmjs.org/entities/-/entities-3.0.1.tgz",
-            "integrity": "sha512-WiyBqoomrwMdFG1e0kqvASYfnlb0lp8M5o5Fw2OFq1hNZxxcNk8Ik0Xm7LxzBhuidnZB/UtBqVCgUz3kBOP51Q==",
-            "dev": true
-        },
-        "fs.realpath": {
-            "version": "1.0.0",
-            "resolved": "https://registry.npmjs.org/fs.realpath/-/fs.realpath-1.0.0.tgz",
-            "integrity": "sha512-OO0pH2lK6a0hZnAdau5ItzHPI6pUlvI7jMVnxUQRtw4owF2wk8lOSabtGDCTP4Ggrg2MbGnWO9X8K1t4+fGMDw==",
-            "dev": true
-        },
-        "get-stdin": {
-            "version": "9.0.0",
-            "resolved": "https://registry.npmjs.org/get-stdin/-/get-stdin-9.0.0.tgz",
-            "integrity": "sha512-dVKBjfWisLAicarI2Sf+JuBE/DghV4UzNAVe9yhEJuzeREd3JhOTE9cUaJTeSa77fsbQUK3pcOpJfM59+VKZaA==",
-            "dev": true
-        },
-        "glob": {
-            "version": "8.0.3",
-            "resolved": "https://registry.npmjs.org/glob/-/glob-8.0.3.tgz",
-            "integrity": "sha512-ull455NHSHI/Y1FqGaaYFaLGkNMMJbavMrEGFXG/PGrg6y7sutWHUHrz6gy6WEBH6akM1M414dWKCNs+IhKdiQ==",
-            "dev": true,
-            "requires": {
-                "fs.realpath": "^1.0.0",
-                "inflight": "^1.0.4",
-                "inherits": "2",
-                "minimatch": "^5.0.1",
-                "once": "^1.3.0"
-            }
-        },
-        "ignore": {
-            "version": "5.2.0",
-            "resolved": "https://registry.npmjs.org/ignore/-/ignore-5.2.0.tgz",
-            "integrity": "sha512-CmxgYGiEPCLhfLnpPp1MoRmifwEIOgjcHXxOBjv7mY96c+eWScsOP9c112ZyLdWHi0FxHjI+4uVhKYp/gcdRmQ==",
-            "dev": true
-        },
-        "inflight": {
-            "version": "1.0.6",
-            "resolved": "https://registry.npmjs.org/inflight/-/inflight-1.0.6.tgz",
-            "integrity": "sha512-k92I/b08q4wvFscXCLvqfsHCrjrF7yiXsQuIVvVE7N82W3+aqpzuUdBbfhWcy/FZR3/4IgflMgKLOsvPDrGCJA==",
-            "dev": true,
-            "requires": {
-                "once": "^1.3.0",
-                "wrappy": "1"
-            }
-        },
-        "inherits": {
-            "version": "2.0.4",
-            "resolved": "https://registry.npmjs.org/inherits/-/inherits-2.0.4.tgz",
-            "integrity": "sha512-k/vGaX4/Yla3WzyMCvTQOXYeIHvqOKtnqBduzTHpzpQZzAskKMhZ2K+EnBiSM9zGSoIFeMpXKxa4dYeZIQqewQ==",
-            "dev": true
-        },
-        "ini": {
-            "version": "3.0.1",
-            "resolved": "https://registry.npmjs.org/ini/-/ini-3.0.1.tgz",
-            "integrity": "sha512-it4HyVAUTKBc6m8e1iXWvXSTdndF7HbdN713+kvLrymxTaU4AUBWrJ4vEooP+V7fexnVD3LKcBshjGGPefSMUQ==",
-            "dev": true
-        },
-        "js-yaml": {
-            "version": "4.1.0",
-            "resolved": "https://registry.npmjs.org/js-yaml/-/js-yaml-4.1.0.tgz",
-            "integrity": "sha512-wpxZs9NoxZaJESJGIZTyDEaYpl0FKSA+FB9aJiyemKhMwkxQg63h4T1KJgUGHpTqPDNRcmmYLugrRjJlBtWvRA==",
-            "dev": true,
-            "requires": {
-                "argparse": "^2.0.1"
-            }
-        },
-        "jsonc-parser": {
-            "version": "3.1.0",
-            "resolved": "https://registry.npmjs.org/jsonc-parser/-/jsonc-parser-3.1.0.tgz",
-            "integrity": "sha512-DRf0QjnNeCUds3xTjKlQQ3DpJD51GvDjJfnxUVWg6PZTo2otSm+slzNAxU/35hF8/oJIKoG9slq30JYOsF2azg==",
-            "dev": true
-        },
-        "linkify-it": {
-            "version": "4.0.1",
-            "resolved": "https://registry.npmjs.org/linkify-it/-/linkify-it-4.0.1.tgz",
-            "integrity": "sha512-C7bfi1UZmoj8+PQx22XyeXCuBlokoyWQL5pWSP+EI6nzRylyThouddufc2c1NDIcP9k5agmN9fLpA7VNJfIiqw==",
-            "dev": true,
-            "requires": {
-                "uc.micro": "^1.0.1"
-            }
-        },
-        "markdown-it": {
-            "version": "13.0.1",
-            "resolved": "https://registry.npmjs.org/markdown-it/-/markdown-it-13.0.1.tgz",
-            "integrity": "sha512-lTlxriVoy2criHP0JKRhO2VDG9c2ypWCsT237eDiLqi09rmbKoUetyGHq2uOIRoRS//kfoJckS0eUzzkDR+k2Q==",
-            "dev": true,
-            "requires": {
-                "argparse": "^2.0.1",
-                "entities": "~3.0.1",
-                "linkify-it": "^4.0.1",
-                "mdurl": "^1.0.1",
-                "uc.micro": "^1.0.5"
-            }
-        },
-        "markdownlint": {
-            "version": "0.26.2",
-            "resolved": "https://registry.npmjs.org/markdownlint/-/markdownlint-0.26.2.tgz",
-            "integrity": "sha512-2Am42YX2Ex5SQhRq35HxYWDfz1NLEOZWWN25nqd2h3AHRKsGRE+Qg1gt1++exW792eXTrR4jCNHfShfWk9Nz8w==",
-            "dev": true,
-            "requires": {
-                "markdown-it": "13.0.1"
-            }
-        },
-        "markdownlint-cli": {
-            "version": "0.32.2",
-            "resolved": "https://registry.npmjs.org/markdownlint-cli/-/markdownlint-cli-0.32.2.tgz",
-            "integrity": "sha512-xmJT1rGueUgT4yGNwk6D0oqQr90UJ7nMyakXtqjgswAkEhYYqjHew9RY8wDbOmh2R270IWjuKSeZzHDEGPAUkQ==",
-            "dev": true,
-            "requires": {
-                "commander": "~9.4.0",
-                "get-stdin": "~9.0.0",
-                "glob": "~8.0.3",
-                "ignore": "~5.2.0",
-                "js-yaml": "^4.1.0",
-                "jsonc-parser": "~3.1.0",
-                "markdownlint": "~0.26.2",
-                "markdownlint-rule-helpers": "~0.17.2",
-                "minimatch": "~5.1.0",
-                "run-con": "~1.2.11"
-            }
-        },
-        "markdownlint-rule-helpers": {
-            "version": "0.17.2",
-            "resolved": "https://registry.npmjs.org/markdownlint-rule-helpers/-/markdownlint-rule-helpers-0.17.2.tgz",
-            "integrity": "sha512-XaeoW2NYSlWxMCZM2B3H7YTG6nlaLfkEZWMBhr4hSPlq9MuY2sy83+Xr89jXOqZMZYjvi5nBCGoFh7hHoPKZmA==",
-            "dev": true
-        },
-        "mdurl": {
-            "version": "1.0.1",
-            "resolved": "https://registry.npmjs.org/mdurl/-/mdurl-1.0.1.tgz",
-            "integrity": "sha512-/sKlQJCBYVY9Ers9hqzKou4H6V5UWc/M59TH2dvkt+84itfnq7uFOMLpOiOS4ujvHP4etln18fmIxA5R5fll0g==",
-            "dev": true
-        },
-        "minimatch": {
-            "version": "5.1.0",
-            "resolved": "https://registry.npmjs.org/minimatch/-/minimatch-5.1.0.tgz",
-            "integrity": "sha512-9TPBGGak4nHfGZsPBohm9AWg6NoT7QTCehS3BIJABslyZbzxfV78QM2Y6+i741OPZIafFAaiiEMh5OyIrJPgtg==",
-            "dev": true,
-            "requires": {
-                "brace-expansion": "^2.0.1"
-            }
-        },
-        "minimist": {
-            "version": "1.2.6",
-            "resolved": "https://registry.npmjs.org/minimist/-/minimist-1.2.6.tgz",
-            "integrity": "sha512-Jsjnk4bw3YJqYzbdyBiNsPWHPfO++UGG749Cxs6peCu5Xg4nrena6OVxOYxrQTqww0Jmwt+Ref8rggumkTLz9Q==",
-            "dev": true
-        },
-        "once": {
-            "version": "1.4.0",
-            "resolved": "https://registry.npmjs.org/once/-/once-1.4.0.tgz",
-            "integrity": "sha512-lNaJgI+2Q5URQBkccEKHTQOPaXdUxnZZElQTZY0MFUAuaEqe1E+Nyvgdz/aIyNi6Z9MzO5dv1H8n58/GELp3+w==",
-            "dev": true,
-            "requires": {
-                "wrappy": "1"
-            }
-        },
-        "pyright": {
-            "version": "1.1.274",
-            "resolved": "https://registry.npmjs.org/pyright/-/pyright-1.1.274.tgz",
-            "integrity": "sha512-+MEiHktoAxlehWbBAF2vwJkJxV5tObbLK6vnc7u+iDBH06vuHqTqY33AV+YGNwfDEU35chw5zu7H3EpGROXsGQ==",
-            "dev": true
-        },
-        "run-con": {
-            "version": "1.2.11",
-            "resolved": "https://registry.npmjs.org/run-con/-/run-con-1.2.11.tgz",
-            "integrity": "sha512-NEMGsUT+cglWkzEr4IFK21P4Jca45HqiAbIIZIBdX5+UZTB24Mb/21iNGgz9xZa8tL6vbW7CXmq7MFN42+VjNQ==",
-            "dev": true,
-            "requires": {
-                "deep-extend": "^0.6.0",
-                "ini": "~3.0.0",
-                "minimist": "^1.2.6",
-                "strip-json-comments": "~3.1.1"
-            }
-        },
-        "strip-json-comments": {
-            "version": "3.1.1",
-            "resolved": "https://registry.npmjs.org/strip-json-comments/-/strip-json-comments-3.1.1.tgz",
-            "integrity": "sha512-6fPc+R4ihwqP6N/aIv2f1gMH8lOVtWQHoqC4yK6oSDVVocumAsfCqjkXnqiYMhmMwS/mEHLp7Vehlt3ql6lEig==",
-            "dev": true
-        },
-        "uc.micro": {
-            "version": "1.0.6",
-            "resolved": "https://registry.npmjs.org/uc.micro/-/uc.micro-1.0.6.tgz",
-            "integrity": "sha512-8Y75pvTYkLJW2hWQHXxoqRgV7qb9B+9vFEtidML+7koHUFapnVJAZ6cKs+Qjz5Aw3aZWHMC6u0wJE3At+nSGwA==",
-            "dev": true
-        },
-        "wrappy": {
-            "version": "1.0.2",
-            "resolved": "https://registry.npmjs.org/wrappy/-/wrappy-1.0.2.tgz",
-            "integrity": "sha512-l4Sp/DRseor9wL6EvV2+TuQn63dMkPjZ/sp9XkghTEbV9KlPS1xUsZ3u7/IQO4wxtcFB4bgpQPRcR3QCvezPcQ==",
-            "dev": true
-        }
     }
 }

package.json

@@ -10,7 +10,7 @@
     "version": "1.0.0",
     "dependencies": {},
     "devDependencies": {
-        "markdownlint-cli": "^0.32.1",
-        "pyright": "^1.1.274"
+        "markdownlint-cli": "^0.33.0",
+        "pyright": "^1.1.301"
     }
 }

pyproject.toml

@@ -1,9 +1,9 @@
 [misc]
-stable_version = "0.5.1"
+stable_version = "0.6.6"
 
 [tool.poetry]
 name = "whispering"
-version = "0.5.1"
+version = "0.6.7pre0"
 description = "Streaming transcriber with whisper"
 license = "MIT"
 authors = ["Yuta Hayashibe <yuta@hayashibe.jp>"]
@@ -13,13 +13,17 @@ packages = [{include = "whispering"}]
 
 [tool.poetry.dependencies]
 python = ">=3.8,<3.11"
-whisper = {git = "https://github.com/openai/whisper.git", rev = 'd18e9ea5dd2ca57c697e8e55f9e654f06ede25d0'}
+whisper = {git = "https://github.com/openai/whisper.git", rev = 'fd8f80c8b880dd7c284c109ca7f03dbe978bc532'}
 sounddevice = "^0.4.5"
-pydantic = "^1.10.2"
-websockets = "^10.3"
+pydantic = "^1.10.4"
+websockets = "^10.4"
 tqdm = "*"
-torchaudio = "^0.12.1"
+torchaudio = "^0.13.1"
+torch = ">=1.13.1"
 PySoundFile = {version = "^0.9.0.post1", platform = "windows"}
+numpy = "^1.24.2"
+ffmpeg-python = "^0.2.0"
+transformers = "^4.26.1"
 
 
 [tool.poetry.group.dev.dependencies]
@@ -28,7 +32,7 @@ isort = ">=5.10.1"
 flake8 = ">=5.0.4"
 pydocstyle = ">=6.1.1"
 toml = "^0.10.2"
-pytest = "^7.1.3"
+pytest = "^7.2.1"
 
 [build-system]
 requires = ["poetry-core>=1.3.2"]

tests/test_cli.py

@@ -1,28 +1,42 @@
 #!/usr/bin/env python3
 
 
-import sys
-from unittest.mock import patch
+from pydantic import BaseModel
 
-from whispering.cli import get_opts, is_valid_arg
+from whispering.cli import Mode, is_valid_arg
+
+
+class ArgExample(BaseModel):
+    mode: Mode
+    cmd: str
+    ok: bool
 
 
 def test_options():
 
-    invalid_args = [
-        "--mode server --mic 0",
-        "--mode server --mic 1",
-        "--mode server --beam_size 3",
-        "--mode server --temperature 0",
-        "--mode server --allow-padding",
-        "--mode server --no-vad",
-        "--mode server --num_block 3",
-        "--mode mic --host 0.0.0.0",
-        "--mode mic --port 8000",
+    exs = [
+        ArgExample(mode=Mode.server, cmd="--mic 0", ok=False),
+        ArgExample(mode=Mode.server, cmd="--mic 1", ok=False),
+        ArgExample(
+            mode=Mode.server,
+            cmd="--host 0.0.0.0 --port 8000",
+            ok=True,
+        ),
+        ArgExample(
+            mode=Mode.server,
+            cmd="--language en --model tiny --host 0.0.0.0 --port 8000",
+            ok=True,
+        ),
+        ArgExample(mode=Mode.server, cmd="--beam_size 3", ok=False),
+        ArgExample(mode=Mode.server, cmd="--temperature 0", ok=False),
+        ArgExample(mode=Mode.server, cmd="--num_block 3", ok=False),
+        ArgExample(mode=Mode.mic, cmd="--host 0.0.0.0", ok=False),
+        ArgExample(mode=Mode.mic, cmd="--port 8000", ok=False),
     ]
 
-    for invalid_arg in invalid_args:
-        with patch.object(sys, "argv", [""] + invalid_arg.split()):
-            opts = get_opts()
-            ok = is_valid_arg(opts)
-            assert ok is False, f"{invalid_arg} should be invalid"
+    for ex in exs:
+        ok = is_valid_arg(
+            mode=ex.mode.value,
+            args=ex.cmd.split(),
+        )
+        assert ok is ex.ok, f"{ex.cmd} should be {ex.ok}"

whispering/cli.py

@@ -7,7 +7,7 @@ import sys
 from enum import Enum
 from logging import DEBUG, INFO, basicConfig, getLogger
 from pathlib import Path
-from typing import Iterator, Optional, Union
+from typing import Iterator, List, Optional, Union
 
 import sounddevice as sd
 import torch
@@ -16,7 +16,12 @@ from whisper.audio import N_FRAMES, SAMPLE_RATE
 from whisper.tokenizer import LANGUAGES, TO_LANGUAGE_CODE
 
 from whispering.pbar import ProgressBar
-from whispering.schema import Context, StdoutWriter, WhisperConfig
+from whispering.schema import (
+    CURRENT_PROTOCOL_VERSION,
+    Context,
+    StdoutWriter,
+    WhisperConfig,
+)
 from whispering.serve import serve_with_websocket
 from whispering.transcriber import WhisperStreamingTranscriber
 from whispering.websocket_client import run_websocket_client
@@ -140,12 +145,22 @@ def get_opts() -> argparse.Namespace:
         default=[],
     )
     group_ctx.add_argument(
-        "--allow-padding",
-        action="store_true",
+        "--vad",
+        type=float,
+        help="Threshold of VAD",
+        default=0.5,
     )
     group_ctx.add_argument(
-        "--no-vad",
-        action="store_true",
+        "--max_nospeech_skip",
+        type=int,
+        help="Maximum number of skip to analyze because of nospeech",
+        default=16,
+    )
+    group_ctx.add_argument(
+        "--frame",
+        type=int,
+        help="The number of minimum frames of mel spectrogram input for Whisper",
+        default=N_FRAMES,
     )
 
     group_misc = parser.add_argument_group("Other options")
@@ -214,10 +229,12 @@ def get_wshiper(*, opts) -> WhisperStreamingTranscriber:
 
 def get_context(*, opts) -> Context:
     ctx = Context(
+        protocol_version=CURRENT_PROTOCOL_VERSION,
         beam_size=opts.beam_size,
         temperatures=opts.temperature,
-        allow_padding=opts.allow_padding,
-        vad=not opts.no_vad,
+        max_nospeech_skip=opts.max_nospeech_skip,
+        vad_threshold=opts.vad,
+        mel_frame_min_num=opts.frame,
     )
     logger.debug(f"Context: {ctx}")
     return ctx
@@ -230,26 +247,36 @@ def show_devices():
             print(f"{i}: {device['name']}")
 
 
-def is_valid_arg(opts) -> bool:
+def is_valid_arg(
+    *,
+    args: List[str],
+    mode: str,
+) -> bool:
     keys = []
-    if opts.mode == Mode.server.value:
-        keys = [
-            "mic",
-            "beam_size",
-            "temperature",
-            "allow_padding",
-            "no-vad",
-        ]
-    elif opts.mode == Mode.mic.value:
-        keys = [
-            "host",
-            "port",
-        ]
+    if mode == Mode.server.value:
+        keys = {
+            "--mic",
+            "--beam_size",
+            "-b",
+            "--temperature",
+            "-t",
+            "--num_block",
+            "-n",
+            "--vad",
+            "--max_nospeech_skip",
+            "--output",
+            "--show-devices",
+            "--no-progress",
+        }
+    elif mode == Mode.mic.value:
+        keys = {
+            "--host",
+            "--port",
+        }
 
-    for key in keys:
-        _val = vars(opts).get(key)
-        if _val is not None and _val is not False:
-            sys.stderr.write(f"{key} is not accepted option for {opts.mode} mode\n")
+    for arg in args:
+        if arg in keys:
+            sys.stderr.write(f"{arg} is not accepted option for {mode} mode\n")
             return False
     return True
 
@@ -272,7 +299,10 @@ def main() -> None:
     ):
         opts.mode = Mode.server.value
 
-    if not is_valid_arg(opts):
+    if not is_valid_arg(
+        args=sys.argv[1:],
+        mode=opts.mode,
+    ):
         sys.exit(1)
 
     if opts.mode == Mode.client.value:

whispering/schema.py

@@ -1,11 +1,12 @@
 #!/usr/bin/env python3
 
 import sys
-from typing import List, Optional
+from typing import Final, List, Optional
 
 import numpy as np
 import torch
-from pydantic import BaseModel, root_validator
+from pydantic import BaseModel, Field, root_validator
+from whisper.audio import N_FRAMES
 
 
 class WhisperConfig(BaseModel):
@@ -24,14 +25,17 @@ class WhisperConfig(BaseModel):
         return values
 
 
+CURRENT_PROTOCOL_VERSION: Final[int] = int("000_006_003")
+
+
 class Context(BaseModel, arbitrary_types_allowed=True):
+    protocol_version: int
     timestamp: float = 0.0
     buffer_tokens: List[torch.Tensor] = []
     buffer_mel: Optional[torch.Tensor] = None
-    vad: bool = True
+    nosoeech_skip_count: Optional[int] = None
 
     temperatures: List[float]
-    allow_padding: bool = False
     patience: Optional[float] = None
     compression_ratio_threshold: Optional[float] = 2.4
     logprob_threshold: Optional[float] = -1.0
@@ -42,7 +46,11 @@ class Context(BaseModel, arbitrary_types_allowed=True):
     logprob_threshold: Optional[float] = -1.0
     compression_ratio_threshold: Optional[float] = 2.4
     buffer_threshold: Optional[float] = 0.5
-    vad_threshold: float = 0.5
+    vad_threshold: float
+    max_nospeech_skip: int
+    mel_frame_min_num: int = Field(N_FRAMES, ge=1, le=N_FRAMES)
+
+    data_type: str = "float32"
 
 
 class ParsedChunk(BaseModel):

whispering/serve.py

@@ -3,16 +3,20 @@
 import asyncio
 import json
 from logging import getLogger
-from typing import Optional
+from typing import Final, Optional
 
 import numpy as np
 import websockets
 from websockets.exceptions import ConnectionClosedOK
 
-from whispering.transcriber import Context, WhisperStreamingTranscriber
+from whispering.schema import CURRENT_PROTOCOL_VERSION, Context
+from whispering.transcriber import WhisperStreamingTranscriber
 
 logger = getLogger(__name__)
 
+MIN_PROTOCOL_VERSION: Final[int] = int("000_006_000")
+MAX_PROTOCOL_VERSION: Final[int] = CURRENT_PROTOCOL_VERSION
+
 
 async def serve_with_websocket_main(websocket):
     global g_wsp
@@ -41,10 +45,28 @@ async def serve_with_websocket_main(websocket):
                     )
                 )
                 return
+
+            if ctx.protocol_version < MIN_PROTOCOL_VERSION:
+                await websocket.send(
+                    json.dumps(
+                        {
+                            "error": f"protocol_version is older than {MIN_PROTOCOL_VERSION}"
+                        }
+                    )
+                )
+            elif ctx.protocol_version > MAX_PROTOCOL_VERSION:
+                await websocket.send(
+                    json.dumps(
+                        {
+                            "error": f"protocol_version is newer than {MAX_PROTOCOL_VERSION}"
+                        }
+                    )
+                )
+                return
+
             continue
 
         logger.debug(f"Message size: {len(message)}")
-        audio = np.frombuffer(message, dtype=np.float32)
         if ctx is None:
             await websocket.send(
                 json.dumps(
@@ -54,6 +76,7 @@ async def serve_with_websocket_main(websocket):
                 )
             )
             return
+        audio = np.frombuffer(message, dtype=np.dtype(ctx.data_type)).astype(np.float32)
         for chunk in g_wsp.transcribe(
             audio=audio,  # type: ignore
             ctx=ctx,

whispering/transcriber.py

@@ -30,9 +30,9 @@ class WhisperStreamingTranscriber:
         self.dtype = torch.float16 if fp16 else torch.float32
         if self.model.device == torch.device("cpu"):
             if torch.cuda.is_available():
-                logger.warning("Performing inference on CPU when CUDA is available")
+                logger.info("Performing inference on CPU though CUDA is available")
             if self.dtype == torch.float16:
-                logger.warning("FP16 is not supported on CPU; using FP32 instead")
+                logger.info("Using FP32 because FP16 is not supported on CPU")
                 self.dtype = torch.float32
 
         if self.dtype == torch.float32:
@@ -233,8 +233,9 @@ class WhisperStreamingTranscriber:
         ctx: Context,
     ) -> Iterator[ParsedChunk]:
         logger.debug(f"{len(audio)}")
+        force_padding: bool = False
 
-        if ctx.vad:
+        if ctx.vad_threshold > 0.0:
             x = [
                 v
                 for v in self.vad(
@@ -246,7 +247,20 @@ class WhisperStreamingTranscriber:
             if len(x) == 0:  # No speech
                 logger.debug("No speech")
                 ctx.timestamp += len(audio) / N_FRAMES * self.duration_pre_one_mel
-                return
+
+                if ctx.nosoeech_skip_count is not None:
+                    ctx.nosoeech_skip_count += 1
+
+                if (
+                    ctx.nosoeech_skip_count is None
+                    or ctx.nosoeech_skip_count <= ctx.max_nospeech_skip
+                ):
+                    logger.debug(
+                        f"nosoeech_skip_count: {ctx.nosoeech_skip_count} (<= {ctx.max_nospeech_skip})"
+                    )
+                    return
+                ctx.nosoeech_skip_count = None
+                force_padding = True
 
         new_mel = log_mel_spectrogram(audio=audio)
         logger.debug(f"Incoming new_mel.shape: {new_mel.shape}")
@@ -261,12 +275,15 @@ class WhisperStreamingTranscriber:
         seek: int = 0
         while seek < mel.shape[-1]:
             logger.debug(f"seek: {seek}")
-            if mel.shape[-1] - seek < N_FRAMES:
+            if mel.shape[-1] - seek <= 0:
+                logger.debug(f"No more seek: mel.shape={mel.shape}, seek={seek}")
+                break
+            if mel.shape[-1] - seek < ctx.mel_frame_min_num:
                 logger.debug(
-                    f"mel.shape ({mel.shape[-1]}) - seek ({seek}) < N_FRAMES ({N_FRAMES})"
+                    f"mel.shape ({mel.shape[-1]}) - seek ({seek}) < ctx.mel_frame_min_num ({ctx.mel_frame_min_num})"
                 )
-                if ctx.allow_padding:
-                    logger.warning("Padding is not expected while speaking")
+                if force_padding:
+                    logger.debug("Padding")
                 else:
                     logger.debug("No padding")
                     break
@@ -319,9 +336,13 @@ class WhisperStreamingTranscriber:
             logger.debug(f"new seek={seek}, mel.shape: {mel.shape}")
 
         if mel.shape[-1] - seek <= 0:
+            ctx.buffer_mel = None
+            ctx.nosoeech_skip_count = None
             logger.debug(f"ctx.buffer_mel is None ({mel.shape}, {seek})")
             return
         ctx.buffer_mel = mel[:, seek:]
         assert ctx.buffer_mel is not None
         logger.debug(f"ctx.buffer_mel.shape: {ctx.buffer_mel.shape}")
         del mel
+        if ctx.nosoeech_skip_count is None:
+            ctx.nosoeech_skip_count = 0  # start count

whispering/vad.py

@@ -1,5 +1,6 @@
 #!/usr/bin/env python3
 
+from logging import getLogger
 from typing import Iterator, Optional
 
 import numpy as np
@@ -8,6 +9,8 @@ from whisper.audio import N_FRAMES, SAMPLE_RATE
 
 from whispering.schema import SpeechSegment
 
+logger = getLogger(__name__)
+
 
 class VAD:
     def __init__(
@@ -50,6 +53,7 @@ class VAD:
                 torch.from_numpy(audio[start:end]),
                 SAMPLE_RATE,
             ).item()
+            logger.debug(f"VAD: {vad_prob} (threshold={threshold})")
             if vad_prob > threshold:
                 if start_block_idx is None:
                     start_block_idx = idx