Skip to content

Commit 86845d6

Browse files
franckfermanfranckferman
committed
Initial commit
0 parents  commit 86845d6

28 files changed

Lines changed: 5764 additions & 0 deletions

‎.gitignore‎

Lines changed: 38 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,38 @@
1+
# Python
2+
__pycache__/
3+
*.py[cod]
4+
*.so
5+
.venv/
6+
7+
# Distribution
8+
build/
9+
dist/
10+
*.egg-info/
11+
12+
# Tests and coverage
13+
.pytest_cache/
14+
.coverage
15+
htmlcov/
16+
17+
# Type checkers
18+
.mypy_cache/
19+
20+
# Editors
21+
.vscode/
22+
.idea/
23+
*~
24+
*.swp
25+
26+
# OS
27+
.DS_Store
28+
Thumbs.db
29+
30+
# Project-specific
31+
config.json
32+
output/
33+
transcript.txt
34+
transcript.srt
35+
transcript.vtt
36+
transcript.json
37+
whisper.cpp/
38+
*.log

‎LICENSE‎

Lines changed: 661 additions & 0 deletions
Large diffs are not rendered by default.

‎Makefile‎

Lines changed: 204 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,204 @@
1+
PYTHON := python3
2+
VENV := .venv
3+
PIP := $(VENV)/bin/pip
4+
PY := $(VENV)/bin/python
5+
6+
# ── Default config/output ─────────────────────────────────────────────────────
7+
CONFIG ?= config.json
8+
BACKEND ?= faster_whisper
9+
LANGUAGE ?= fr
10+
WORKERS ?= 2
11+
OUTPUT ?= ./output
12+
FILE ?=
13+
URL ?=
14+
15+
.DEFAULT_GOAL := help
16+
17+
# ── Setup ─────────────────────────────────────────────────────────────────────
18+
19+
.PHONY: install
20+
install: ## Full installation (venv + deps + config)
21+
bash install.sh
22+
23+
.PHONY: venv
24+
venv: ## Create virtual environment only
25+
$(PYTHON) -m venv $(VENV)
26+
$(PIP) install --upgrade pip --quiet
27+
28+
.PHONY: deps
29+
deps: venv ## Install core dependencies
30+
$(PIP) install --quiet ffmpeg-python tqdm requests yt-dlp pydub
31+
32+
.PHONY: deps-faster-whisper
33+
deps-faster-whisper: deps ## Install faster-whisper backend
34+
$(PIP) install --quiet faster-whisper
35+
36+
.PHONY: deps-openai
37+
deps-openai: deps ## Install OpenAI backend
38+
$(PIP) install --quiet openai
39+
40+
.PHONY: deps-all
41+
deps-all: deps deps-faster-whisper deps-openai ## Install all Python dependencies
42+
43+
# ── whisper.cpp ───────────────────────────────────────────────────────────────
44+
45+
WHISPER_CPP_DIR := whisper.cpp
46+
WHISPER_MODEL ?= base
47+
WHISPER_MODEL_BIN := $(WHISPER_CPP_DIR)/models/ggml-$(WHISPER_MODEL).bin
48+
49+
.PHONY: whisper-cpp-build
50+
whisper-cpp-build: ## Clone and compile whisper.cpp
51+
@if [ ! -d "$(WHISPER_CPP_DIR)" ]; then \
52+
echo "[+] Cloning whisper.cpp..."; \
53+
git clone --depth=1 https://github.com/ggerganov/whisper.cpp $(WHISPER_CPP_DIR); \
54+
else \
55+
echo "[+] whisper.cpp already cloned."; \
56+
fi
57+
@echo "[+] Compiling..."
58+
$(MAKE) -C $(WHISPER_CPP_DIR) -j$$(nproc)
59+
@echo "[+] Build complete."
60+
@ls $(WHISPER_CPP_DIR)/main $(WHISPER_CPP_DIR)/build/bin/whisper-cli 2>/dev/null | head -1 | xargs -I{} echo "[+] Binary: {}"
61+
62+
.PHONY: whisper-cpp-model
63+
whisper-cpp-model: ## Download whisper.cpp model (WHISPER_MODEL=base|tiny|small|medium|large-v2)
64+
@if [ -f "$(WHISPER_MODEL_BIN)" ]; then \
65+
echo "[+] Model already exists: $(WHISPER_MODEL_BIN)"; \
66+
else \
67+
echo "[+] Downloading model: $(WHISPER_MODEL)..."; \
68+
bash $(WHISPER_CPP_DIR)/models/download-ggml-model.sh $(WHISPER_MODEL); \
69+
fi
70+
71+
.PHONY: whisper-cpp-setup
72+
whisper-cpp-setup: whisper-cpp-build whisper-cpp-model ## Full whisper.cpp setup (build + model)
73+
74+
# ── Run ───────────────────────────────────────────────────────────────────────
75+
76+
.PHONY: run
77+
run: ## Transcribe FILE= or URL= (e.g. make run FILE=video.mp4)
78+
ifndef FILE
79+
ifndef URL
80+
$(error Specify FILE=path/to/video.mp4 or URL=https://...)
81+
endif
82+
endif
83+
ifdef FILE
84+
$(PY) main.py --file "$(FILE)" --backend $(BACKEND) --language $(LANGUAGE) \
85+
--workers $(WORKERS) --output-dir $(OUTPUT)
86+
endif
87+
ifdef URL
88+
$(PY) main.py --url "$(URL)" --backend $(BACKEND) --language $(LANGUAGE) \
89+
--workers $(WORKERS) --output-dir $(OUTPUT)
90+
endif
91+
92+
.PHONY: run-config
93+
run-config: ## Run using CONFIG= file (default: config.json)
94+
$(PY) main.py --config $(CONFIG)
95+
96+
.PHONY: dry-run
97+
dry-run: ## Dry-run — show what would be executed without running
98+
ifdef FILE
99+
$(PY) main.py --file "$(FILE)" --backend $(BACKEND) --dry-run
100+
else ifdef URL
101+
$(PY) main.py --url "$(URL)" --backend $(BACKEND) --dry-run
102+
else
103+
$(error Specify FILE= or URL=)
104+
endif
105+
106+
.PHONY: run-whisper-cpp
107+
run-whisper-cpp: ## Run with whisper_cpp backend (auto-detects binary and model)
108+
$(eval WCPP_BIN := $(shell ls $(WHISPER_CPP_DIR)/build/bin/whisper-cli $(WHISPER_CPP_DIR)/main 2>/dev/null | head -1))
109+
@if [ -z "$(WCPP_BIN)" ]; then echo "[x] whisper.cpp binary not found. Run: make whisper-cpp-setup"; exit 1; fi
110+
@if [ ! -f "$(WHISPER_MODEL_BIN)" ]; then echo "[x] Model not found: $(WHISPER_MODEL_BIN). Run: make whisper-cpp-model"; exit 1; fi
111+
ifdef FILE
112+
$(PY) main.py --file "$(FILE)" --backend whisper_cpp \
113+
--whisper-binary "$(WCPP_BIN)" --whisper-model "$(WHISPER_MODEL_BIN)" \
114+
--language $(LANGUAGE) --workers $(WORKERS) --output-dir $(OUTPUT)
115+
else ifdef URL
116+
$(PY) main.py --url "$(URL)" --backend whisper_cpp \
117+
--whisper-binary "$(WCPP_BIN)" --whisper-model "$(WHISPER_MODEL_BIN)" \
118+
--language $(LANGUAGE) --workers $(WORKERS) --output-dir $(OUTPUT)
119+
else
120+
$(error Specify FILE= or URL=)
121+
endif
122+
123+
.PHONY: run-faster-whisper
124+
run-faster-whisper: ## Run with faster_whisper backend
125+
$(MAKE) run BACKEND=faster_whisper
126+
127+
.PHONY: run-openai
128+
run-openai: ## Run with OpenAI API backend
129+
$(MAKE) run BACKEND=openai
130+
131+
# ── Output formats ────────────────────────────────────────────────────────────
132+
133+
.PHONY: run-srt
134+
run-srt: ## Transcribe and output SRT subtitles
135+
ifdef FILE
136+
$(PY) main.py --file "$(FILE)" --backend $(BACKEND) --language $(LANGUAGE) \
137+
--format srt --output-dir $(OUTPUT)
138+
else
139+
$(error Specify FILE=path/to/video.mp4)
140+
endif
141+
142+
.PHONY: run-all-formats
143+
run-all-formats: ## Transcribe and output txt + json + srt + vtt
144+
ifdef FILE
145+
$(PY) main.py --file "$(FILE)" --backend $(BACKEND) --language $(LANGUAGE) \
146+
--format txt,json,srt,vtt --output-dir $(OUTPUT)
147+
else
148+
$(error Specify FILE=path/to/video.mp4)
149+
endif
150+
151+
# ── Dev ───────────────────────────────────────────────────────────────────────
152+
153+
.PHONY: check
154+
check: ## Validate all imports and CLI
155+
$(PY) -c "from transcriber.config import TranscriptionConfig; print('config OK')"
156+
$(PY) -c "from transcriber.backends.base import TranscriptionBackend; print('backends OK')"
157+
$(PY) -c "from transcriber.managers.transcription import TranscriptionManager; print('manager OK')"
158+
$(PY) -c "from transcriber.formatters.output import OutputFormatter; print('formatter OK')"
159+
$(PY) main.py --help > /dev/null && echo "CLI OK"
160+
161+
.PHONY: clean
162+
clean: ## Remove output files and temp dirs
163+
rm -rf output/* __pycache__ transcriber/__pycache__
164+
find . -type d -name "__pycache__" -exec rm -rf {} + 2>/dev/null || true
165+
find . -name "*.pyc" -delete
166+
167+
.PHONY: clean-all
168+
clean-all: clean ## Remove venv and output
169+
rm -rf $(VENV)
170+
171+
# ── Web interface ─────────────────────────────────────────────────────────────
172+
173+
.PHONY: web
174+
web: ## Start web interface at http://localhost:8000
175+
$(PIP) install --quiet fastapi "uvicorn[standard]" python-multipart
176+
$(PY) -m uvicorn web.app:app --host 0.0.0.0 --port 8000 --reload
177+
178+
.PHONY: web-install
179+
web-install: ## Install web dependencies only
180+
$(PIP) install --quiet fastapi "uvicorn[standard]" python-multipart
181+
182+
# ── Help ──────────────────────────────────────────────────────────────────────
183+
184+
.PHONY: help
185+
help: ## Show this help
186+
@echo ""
187+
@echo " Whisper Transcriber"
188+
@echo ""
189+
@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) \
190+
| awk 'BEGIN {FS = ":.*?## "}; {printf " \033[36m%-22s\033[0m %s\n", $$1, $$2}'
191+
@echo ""
192+
@echo " Variables (override with make VAR=value):"
193+
@echo " BACKEND $(BACKEND)"
194+
@echo " LANGUAGE $(LANGUAGE)"
195+
@echo " WORKERS $(WORKERS)"
196+
@echo " OUTPUT $(OUTPUT)"
197+
@echo ""
198+
@echo " Examples:"
199+
@echo " make run FILE=video.mp4"
200+
@echo " make run URL=https://youtube.com/... BACKEND=whisper_cpp LANGUAGE=en"
201+
@echo " make run-all-formats FILE=video.mp4"
202+
@echo " make dry-run URL=https://..."
203+
@echo " make check"
204+
@echo ""

0 commit comments

Comments
 (0)