Commit @d0feed84e876122a18a648512f30adc5eddd1f6a - yjyoon/whisper_server

aa5390b

d0feed8

.pre-commit-config.yaml

--- .pre-commit-config.yaml

+++ .pre-commit-config.yaml


       - id: end-of-file-fixer
       - id: check-yaml
       - id: check-added-large-files
-  # TODO: enable
-  # - repo: https://github.com/pre-commit/mirrors-mypy
-  #   rev: v1.10.0
-  #   hooks:
-  #     - id: mypy
-  #       args: [--strict]
-  # TODO: enable
-  # - repo: https://github.com/RobertCraigie/pyright-python
-  #   rev: v1.1.363
-  #   hooks:
-  #   - id: pyright
-  # Disabled because it doesn't work on NixOS
-  # - repo: https://github.com/astral-sh/ruff-pre-commit
-  #   rev: v0.4.4
-  #   hooks:
-  #     - id: ruff # linter
-  #       args: [--fix]
-  #     - id: ruff-format
+  - repo: https://github.com/python-jsonschema/check-jsonschema
+    rev: 0.28.4
+    hooks:
+      - id: check-taskfile
+  - repo: https://github.com/rhysd/actionlint
+    rev: v1.7.0
+    hooks:
+      - id: actionlint
+  - repo: https://github.com/IamTheFij/docker-pre-commit
+    rev: v3.0.1
+    hooks:
+      - id: docker-compose-check
+  - repo: https://github.com/hadolint/hadolint
+    rev: v2.12.0
+    hooks:
+      - id: hadolint
+  - repo: https://github.com/shellcheck-py/shellcheck-py
+    rev: v0.10.0.1
+    hooks:
+      - id: shellcheck
+  # NOTE: not using https://github.com/RobertCraigie/pyright-python because it doesn't work with poetry virtual environments
+  # NOTE: not using github.com/astral-sh/ruff-pre-commit because it doesn't work on NixOS
+  - repo: local
+    hooks:
+      - id: pyright
+        name: pyright
+        entry: ./pre-commit-scripts/pyright.sh
+        language: script
+        pass_filenames: false
+      - id: ruff-lint
+        name: ruff-lint
+        entry: ./pre-commit-scripts/ruff-lint.sh
+        pass_filenames: false
+        language: script
+      - id: ruff-format
+        name: ruff-format
+        entry: ./pre-commit-scripts/ruff-format.sh
+        language: script

aa5390b

d0feed8

Dockerfile.cpu

--- Dockerfile.cpu

+++ Dockerfile.cpu


 FROM ubuntu:22.04
+# hadolint ignore=DL3008,DL4006
 RUN apt-get update && \
-    apt-get install -y curl software-properties-common && \
+    apt-get install -y --no-install-recommends curl software-properties-common && \
     add-apt-repository ppa:deadsnakes/ppa && \
     apt-get update && \
-    DEBIAN_FRONTEND=noninteractive apt-get -y install python3.11 python3.11-distutils && \
+    DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends python3.11 python3.11-distutils && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/* && \
     curl -sS https://bootstrap.pypa.io/get-pip.py | python3.11
 RUN pip install --no-cache-dir poetry==1.8.2
 WORKDIR /root/speaches

aa5390b

d0feed8

Dockerfile.cuda

--- Dockerfile.cuda

+++ Dockerfile.cuda


 FROM nvidia/cuda:12.2.2-cudnn8-runtime-ubuntu22.04
+# hadolint ignore=DL3008,DL4006
 RUN apt-get update && \
-    apt-get install -y curl software-properties-common && \
+    apt-get install -y --no-install-recommends curl software-properties-common && \
     add-apt-repository ppa:deadsnakes/ppa && \
     apt-get update && \
-    DEBIAN_FRONTEND=noninteractive apt-get -y install python3.11 python3.11-distutils && \
+    DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends python3.11 python3.11-distutils && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/* && \
     curl -sS https://bootstrap.pypa.io/get-pip.py | python3.11
 RUN pip install --no-cache-dir poetry==1.8.2
 WORKDIR /root/speaches

d0feed8

pre-commit-scripts/pyright.sh (added)

+++ pre-commit-scripts/pyright.sh

...	...	@@ -0,0 +1,4 @@
	1	+#!/usr/bin/env bash
	2	+# shellcheck disable=SC1091
	3	+source "$(poetry env info --path)"/bin/activate
	4	+pyright

d0feed8

pre-commit-scripts/ruff-format.sh (added)

+++ pre-commit-scripts/ruff-format.sh

...	...	@@ -0,0 +1,2 @@
	1	+#!/usr/bin/env bash
	2	+ruff format

d0feed8

pre-commit-scripts/ruff-lint.sh (added)

+++ pre-commit-scripts/ruff-lint.sh

...	...	@@ -0,0 +1,2 @@
	1	+#!/usr/bin/env bash
	2	+ruff check --fix

aa5390b

d0feed8

pyproject.toml

--- pyproject.toml

+++ pyproject.toml


 [tool.ruff]
 target-version = "py311"
 
+[tool.pyright]
+# typeCheckingMode = "strict"
+pythonVersion = "3.11"
+pythonPlatform = "Linux"
+
 [build-system]
 requires = ["poetry-core"]
 build-backend = "poetry.core.masonry.api"

aa5390b

d0feed8

speaches/main.py

--- speaches/main.py

+++ speaches/main.py


 from io import BytesIO
 from typing import Annotated, Literal, OrderedDict
 
-from fastapi import (FastAPI, Form, Query, Response, UploadFile, WebSocket,
-                     WebSocketDisconnect)
+from fastapi import (
+    FastAPI,
+    Form,
+    Query,
+    Response,
+    UploadFile,
+    WebSocket,
+    WebSocketDisconnect,
+)
 from fastapi.responses import StreamingResponse
 from fastapi.websockets import WebSocketState
 from faster_whisper import WhisperModel

 from speaches import utils
 from speaches.asr import FasterWhisperASR
 from speaches.audio import AudioStream, audio_samples_from_file
-from speaches.config import (SAMPLES_PER_SECOND, Language, Model,
-                             ResponseFormat, config)
+from speaches.config import SAMPLES_PER_SECOND, Language, Model, ResponseFormat, config
 from speaches.logger import logger
-from speaches.server_models import (TranscriptionJsonResponse,
-                                    TranscriptionVerboseJsonResponse)
+from speaches.server_models import (
+    TranscriptionJsonResponse,
+    TranscriptionVerboseJsonResponse,
+)
 from speaches.transcriber import audio_transcriber
 
 models: OrderedDict[Model, WhisperModel] = OrderedDict()

aa5390b

d0feed8

speaches/server_models.py

--- speaches/server_models.py

+++ speaches/server_models.py


             text=segment.text,
             words=(
                 [WordObject.from_word(word) for word in segment.words]
-                if type(segment.words) == list
+                if isinstance(segment.words, list)
                 else []
             ),
             segments=[SegmentObject.from_segment(segment)],

aa5390b

d0feed8

tests/app_test.py

--- tests/app_test.py

+++ tests/app_test.py


 
 from speaches.config import BYTES_PER_SECOND
 from speaches.main import app
-from speaches.server_models import TranscriptionVerboseResponse
+from speaches.server_models import TranscriptionVerboseJsonResponse
 
 SIMILARITY_THRESHOLD = 0.97
 AUDIO_FILES_LIMIT = 5

 
 def transcribe_audio_data(
     client: TestClient, data: bytes
-) -> TranscriptionVerboseResponse:
+) -> TranscriptionVerboseJsonResponse:
     response = client.post(
         TRANSCRIBE_ENDPOINT,
         files={"file": ("audio.raw", data, "audio/raw")},
     )
     data = json.loads(response.json())  # TODO: figure this out
-    return TranscriptionVerboseResponse(**data)  # type: ignore
+    return TranscriptionVerboseJsonResponse(**data)  # type: ignore
 
 
 @pytest.mark.parametrize("file_path", file_paths)

     with open(file_path, "rb") as file:
         data = file.read()
 
-    streaming_transcription: TranscriptionVerboseResponse = None  # type: ignore
+    streaming_transcription: TranscriptionVerboseJsonResponse = None  # type: ignore
     thread = threading.Thread(
         target=stream_audio_data, args=(ws, data), kwargs={"speed": 4.0}
     )
     thread.start()
     while True:
         try:
-            streaming_transcription = TranscriptionVerboseResponse(**ws.receive_json())
+            streaming_transcription = TranscriptionVerboseJsonResponse(
+                **ws.receive_json()
+            )
         except WebSocketDisconnect:
             break
     file_transcription = transcribe_audio_data(client, data)

...	...	@@ -12,7 +12,7 @@
12	12
13	13	from speaches.config import BYTES_PER_SECOND
14	14	from speaches.main import app
15		-from speaches.server_models import TranscriptionVerboseResponse
	15	+from speaches.server_models import TranscriptionVerboseJsonResponse
16	16
17	17	SIMILARITY_THRESHOLD = 0.97
18	18	AUDIO_FILES_LIMIT = 5
...	...	@@ -54,13 +54,13 @@
54	54
55	55	def transcribe_audio_data(
56	56	client: TestClient, data: bytes
57		-) -> TranscriptionVerboseResponse:
	57	+) -> TranscriptionVerboseJsonResponse:
58	58	response = client.post(
59	59	TRANSCRIBE_ENDPOINT,
60	60	files={"file": ("audio.raw", data, "audio/raw")},
61	61	)
62	62	data = json.loads(response.json()) # TODO: figure this out
63		- return TranscriptionVerboseResponse(**data) # type: ignore
	63	+ return TranscriptionVerboseJsonResponse(**data) # type: ignore
64	64
65	65
66	66	@pytest.mark.parametrize("file_path", file_paths)
...	...	@@ -70,14 +70,16 @@
70	70	with open(file_path, "rb") as file:
71	71	data = file.read()
72	72
73		- streaming_transcription: TranscriptionVerboseResponse = None # type: ignore
	73	+ streaming_transcription: TranscriptionVerboseJsonResponse = None # type: ignore
74	74	thread = threading.Thread(
75	75	target=stream_audio_data, args=(ws, data), kwargs={"speed": 4.0}
76	76	)
77	77	thread.start()
78	78	while True:
79	79	try:
80		- streaming_transcription = TranscriptionVerboseResponse(**ws.receive_json())
	80	+ streaming_transcription = TranscriptionVerboseJsonResponse(
	81	+ **ws.receive_json()
	82	+ )
81	83	except WebSocketDisconnect:
82	84	break
83	85	file_transcription = transcribe_audio_data(client, data)

...	...	@@ -8,21 +8,41 @@
8	8	- id: end-of-file-fixer
9	9	- id: check-yaml
10	10	- id: check-added-large-files
11		- # TODO: enable
12		- # - repo: https://github.com/pre-commit/mirrors-mypy
13		- # rev: v1.10.0
14		- # hooks:
15		- # - id: mypy
16		- # args: [--strict]
17		- # TODO: enable
18		- # - repo: https://github.com/RobertCraigie/pyright-python
19		- # rev: v1.1.363
20		- # hooks:
21		- # - id: pyright
22		- # Disabled because it doesn't work on NixOS
23		- # - repo: https://github.com/astral-sh/ruff-pre-commit
24		- # rev: v0.4.4
25		- # hooks:
26		- # - id: ruff # linter
27		- # args: [--fix]
28		- # - id: ruff-format
	11	+ - repo: https://github.com/python-jsonschema/check-jsonschema
	12	+ rev: 0.28.4
	13	+ hooks:
	14	+ - id: check-taskfile
	15	+ - repo: https://github.com/rhysd/actionlint
	16	+ rev: v1.7.0
	17	+ hooks:
	18	+ - id: actionlint
	19	+ - repo: https://github.com/IamTheFij/docker-pre-commit
	20	+ rev: v3.0.1
	21	+ hooks:
	22	+ - id: docker-compose-check
	23	+ - repo: https://github.com/hadolint/hadolint
	24	+ rev: v2.12.0
	25	+ hooks:
	26	+ - id: hadolint
	27	+ - repo: https://github.com/shellcheck-py/shellcheck-py
	28	+ rev: v0.10.0.1
	29	+ hooks:
	30	+ - id: shellcheck
	31	+ # NOTE: not using https://github.com/RobertCraigie/pyright-python because it doesn't work with poetry virtual environments
	32	+ # NOTE: not using github.com/astral-sh/ruff-pre-commit because it doesn't work on NixOS
	33	+ - repo: local
	34	+ hooks:
	35	+ - id: pyright
	36	+ name: pyright
	37	+ entry: ./pre-commit-scripts/pyright.sh
	38	+ language: script
	39	+ pass_filenames: false
	40	+ - id: ruff-lint
	41	+ name: ruff-lint
	42	+ entry: ./pre-commit-scripts/ruff-lint.sh
	43	+ pass_filenames: false
	44	+ language: script
	45	+ - id: ruff-format
	46	+ name: ruff-format
	47	+ entry: ./pre-commit-scripts/ruff-format.sh
	48	+ language: script

...	...	@@ -1,9 +1,12 @@
1	1	FROM ubuntu:22.04
	2	+# hadolint ignore=DL3008,DL4006
2	3	RUN apt-get update && \
3		- apt-get install -y curl software-properties-common && \
	4	+ apt-get install -y --no-install-recommends curl software-properties-common && \
4	5	add-apt-repository ppa:deadsnakes/ppa && \
5	6	apt-get update && \
6		- DEBIAN_FRONTEND=noninteractive apt-get -y install python3.11 python3.11-distutils && \
	7	+ DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends python3.11 python3.11-distutils && \
	8	+ apt-get clean && \
	9	+ rm -rf /var/lib/apt/lists/* && \
7	10	curl -sS https://bootstrap.pypa.io/get-pip.py \| python3.11
8	11	RUN pip install --no-cache-dir poetry==1.8.2
9	12	WORKDIR /root/speaches

...	...	@@ -1,9 +1,12 @@
1	1	FROM nvidia/cuda:12.2.2-cudnn8-runtime-ubuntu22.04
	2	+# hadolint ignore=DL3008,DL4006
2	3	RUN apt-get update && \
3		- apt-get install -y curl software-properties-common && \
	4	+ apt-get install -y --no-install-recommends curl software-properties-common && \
4	5	add-apt-repository ppa:deadsnakes/ppa && \
5	6	apt-get update && \
6		- DEBIAN_FRONTEND=noninteractive apt-get -y install python3.11 python3.11-distutils && \
	7	+ DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends python3.11 python3.11-distutils && \
	8	+ apt-get clean && \
	9	+ rm -rf /var/lib/apt/lists/* && \
7	10	curl -sS https://bootstrap.pypa.io/get-pip.py \| python3.11
8	11	RUN pip install --no-cache-dir poetry==1.8.2
9	12	WORKDIR /root/speaches

...	...	@@ -22,6 +22,11 @@
22	22	[tool.ruff]
23	23	target-version = "py311"
24	24
	25	+[tool.pyright]
	26	+# typeCheckingMode = "strict"
	27	+pythonVersion = "3.11"
	28	+pythonPlatform = "Linux"
	29	+
25	30	[build-system]
26	31	requires = ["poetry-core"]
27	32	build-backend = "poetry.core.masonry.api"

...	...	@@ -6,8 +6,15 @@
6	6	from io import BytesIO
7	7	from typing import Annotated, Literal, OrderedDict
8	8
9		-from fastapi import (FastAPI, Form, Query, Response, UploadFile, WebSocket,
10		- WebSocketDisconnect)
	9	+from fastapi import (
	10	+ FastAPI,
	11	+ Form,
	12	+ Query,
	13	+ Response,
	14	+ UploadFile,
	15	+ WebSocket,
	16	+ WebSocketDisconnect,
	17	+)
11	18	from fastapi.responses import StreamingResponse
12	19	from fastapi.websockets import WebSocketState
13	20	from faster_whisper import WhisperModel
...	...	@@ -16,11 +23,12 @@
16	23	from speaches import utils
17	24	from speaches.asr import FasterWhisperASR
18	25	from speaches.audio import AudioStream, audio_samples_from_file
19		-from speaches.config import (SAMPLES_PER_SECOND, Language, Model,
20		- ResponseFormat, config)
	26	+from speaches.config import SAMPLES_PER_SECOND, Language, Model, ResponseFormat, config
21	27	from speaches.logger import logger
22		-from speaches.server_models import (TranscriptionJsonResponse,
23		- TranscriptionVerboseJsonResponse)
	28	+from speaches.server_models import (
	29	+ TranscriptionJsonResponse,
	30	+ TranscriptionVerboseJsonResponse,
	31	+)
24	32	from speaches.transcriber import audio_transcriber
25	33
26	34	models: OrderedDict[Model, WhisperModel] = OrderedDict()

...	...	@@ -85,7 +85,7 @@
85	85	text=segment.text,
86	86	words=(
87	87	[WordObject.from_word(word) for word in segment.words]
88		- if type(segment.words) == list
	88	+ if isinstance(segment.words, list)
89	89	else []
90	90	),
91	91	segments=[SegmentObject.from_segment(segment)],

Delete comment