Only show "diarize" option when HF_TOKEN is supplied

Fixed Docker caching issues
Updated Readme

Dockerfile

@@ -25,7 +25,7 @@ COPY --from=swagger-ui /usr/share/nginx/html/swagger-ui-bundle.js swagger-ui-ass
 RUN poetry config virtualenvs.in-project true
 RUN poetry install
 
-RUN $POETRY_VENV/bin/pip install pandas transformers nltk pyannote.audio
+RUN /app/.venv/bin/pip install pandas transformers nltk pyannote.audio
 RUN git clone --depth 1 https://github.com/m-bain/whisperX.git \
     && cd whisperX \
     && $POETRY_VENV/bin/pip install -e .

Dockerfile.gpu

@@ -11,7 +11,7 @@ RUN export DEBIAN_FRONTEND=noninteractive \
     python${PYTHON_VERSION}-venv \
     python3-pip \
     ffmpeg \
-    git \    
+    git \ 
     && rm -rf /var/lib/apt/lists/*
 
 RUN ln -s -f /usr/bin/python${PYTHON_VERSION} /usr/bin/python3 && \
@@ -36,7 +36,7 @@ COPY --from=swagger-ui /usr/share/nginx/html/swagger-ui.css swagger-ui-assets/sw
 COPY --from=swagger-ui /usr/share/nginx/html/swagger-ui-bundle.js swagger-ui-assets/swagger-ui-bundle.js
 
 RUN poetry install
-RUN $POETRY_VENV/bin/pip install torch torchaudio pandas transformers nltk pyannote.audio \
+RUN /app/.venv/bin/pip install torch torchaudio pandas transformers nltk pyannote.audio \
     --index-url https://download.pytorch.org/whl/cu118 \
     --index-url https://pypi.org/simple/
 
@@ -45,4 +45,4 @@ RUN git clone --depth 1 https://github.com/m-bain/whisperX.git \
     && $POETRY_VENV/bin/pip install --no-dependencies -e .
 
 EXPOSE 9000
-CMD gunicorn --bind 0.0.0.0:9000 --workers 1 --timeout 0 app.webservice:app -k uvicorn.workers.UvicornWorker
+CMD gunicorn --bind 0.0.0.0:9000 --workers 1 --timeout 0 app.webservice:app -k uvicorn.workers.UvicornWorker

app/webservice.py

@@ -12,6 +12,8 @@
 from whisper import tokenizer
 
 ASR_ENGINE = os.getenv("ASR_ENGINE", "openai_whisper")
+HF_TOKEN = os.getenv("HF_TOKEN", "")
+
 if ASR_ENGINE == "faster_whisper":
     from .faster_whisper.core import transcribe, language_detection
 elif ASR_ENGINE == "whisperx":
@@ -76,7 +78,7 @@ def asr(
     diarize : bool = Query(
         default=False,
         description="Diarize the input",
-        include_in_schema=(True if ASR_ENGINE == "whisperx" else False)),
+        include_in_schema=(True if ASR_ENGINE == "whisperx" and HF_TOKEN != "" else False)),
     min_speakers : Union[int, None] = Query(
         default=None,
         description="Min speakers in this file",

docs/index.md

@@ -5,6 +5,7 @@ Current release (v1.2.0) supports following whisper models:
 
 - [openai/whisper](https://github.com/openai/whisper)@[v20230918](https://github.com/openai/whisper/releases/tag/v20230918)
 - [guillaumekln/faster-whisper](https://github.com/guillaumekln/faster-whisper)@[0.9.0](https://github.com/guillaumekln/faster-whisper/releases/tag/v0.9.0)
+- [whisperX](https://github.com/m-bain/whisperX)@[v3.1.1](https://github.com/m-bain/whisperX/releases/tag/v3.1.1)
 
 ## Quick Usage