fixes to the notebook

MahmoudAshraf97 · Aug 3, 2023 · 4160c1c · 4160c1c
1 parent 9a230c3
commit 4160c1c
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 6 deletions.
diff --git a/Whisper_Transcription_+_NeMo_Diarization.ipynb b/Whisper_Transcription_+_NeMo_Diarization.ipynb
@@ -29,13 +29,14 @@
       },
       "outputs": [],
       "source": [
-        "!pip install nemo_toolkit[asr]==1.17.0\n",
-        "!pip install faster-whisper==0.5.1\n",
         "!pip install git+https://github.com/m-bain/whisperX.git@4cb167a225c0ebaea127fd6049abfaa3af9f8bb4\n",
+        "!pip install --no-build-isolation nemo_toolkit[asr]==1.19.1 \n",
+        "!pip install faster-whisper==0.7.1\n",
         "!pip install git+https://github.com/facebookresearch/demucs#egg=demucs\n",
-        "!pip install transformers>=4.26.1\n",
         "!pip install deepmultilingualpunctuation\n",
-        "!pip install wget"
+        "!pip install wget\n",
+        "!pip install torch torchaudio torchvision\n",
+        "!pip uninstall -y nvidia-cudnn-cu11"
       ]
     },
     {
@@ -470,7 +471,7 @@
         "        vocal_target = audio_path\n",
         "    else:\n",
         "        vocal_target = os.path.join(\n",
-        "            \"temp_outputs\", \"htdemucs\", os.path.basename(args.audio[:-4]), \"vocals.wav\"\n",
+        "            \"temp_outputs\", \"htdemucs\", os.path.basename(audio_path[:-4]), \"vocals.wav\"\n",
         "        )\n",
         "else:\n",
         "    vocal_target = audio_path"

diff --git a/requirements.txt b/requirements.txt
@@ -2,6 +2,6 @@ wget
 nemo_toolkit[asr]==1.19.1
 transformers>=4.26.1
 faster-whisper==0.7.1
-whisperx==3.1.1
+git+https://github.com/m-bain/whisperX.git@4cb167a225c0ebaea127fd6049abfaa3af9f8bb4
 git+https://github.com/facebookresearch/demucs#egg=demucs
 deepmultilingualpunctuation