Merge branch 'main' into v3

.gitignore

@@ -1,2 +1,3 @@
 whisperx.egg-info/
-**/__pycache__/
+**/__pycache__/
+.ipynb_checkpoints

Dockerfile

@@ -0,0 +1,19 @@
+FROM pytorch/pytorch:1.11.0-cuda11.3-cudnn8-devel
+RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/3bf863cc.pub && \
+    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64/7fa2af80.pub
+RUN apt-get update && \
+    apt-get install -y wget && \
+    wget -qO - https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub | apt-key add - && \
+    apt-get update && \
+    apt-get install -y git && \
+    apt-get install libsndfile1 -y && \
+    apt-get clean
+    
+RUN pip install --upgrade pip
+RUN pip install --upgrade setuptools
+RUN pip install git+https://github.com/m-bain/whisperx.git
+RUN pip install jupyter ipykernel
+EXPOSE 8888
+# Use external volume for data
+ENV NVIDIA_VISIBLE_DEVICES 1
+CMD ["jupyter", "notebook", "--ip=0.0.0.0", "--port=8888", "--NotebookApp.token=''","--NotebookApp.password=''", "--allow-root"]

README.md

@@ -52,6 +52,12 @@ This repository provides fast automatic speaker recognition (70x realtime with l
 
 **Speaker Diarization** is the process of partitioning an audio stream containing human speech into homogeneous segments according to the identity of each speaker.
 
+- v3 pre-release [this branch](https://github.com/m-bain/whisperX/tree/v3) *70x speed-up open-sourced. Using batched whisper with faster-whisper backend*!
+- v2 released, code cleanup, imports whisper library. VAD filtering is now turned on by default, as in the paper.
+- Paper drop🎓👨‍🏫! Please see our [ArxiV preprint](https://arxiv.org/abs/2303.00747) for benchmarking and details of WhisperX. We also introduce more efficient batch inference resulting in large-v2 with *60-70x REAL TIME speed (not provided in this repo).
+- VAD filtering: Voice Activity Detection (VAD) from [Pyannote.audio](https://huggingface.co/pyannote/voice-activity-detection) is used as a preprocessing step to remove reliance on whisper timestamps and only transcribe audio segments containing speech. add `--vad_filter True` flag, increases timestamp accuracy and robustness (requires more GPU mem due to 30s inputs in wav2vec2)
+- Character level timestamps (see `*.char.ass` file output)
+- Diarization (still in beta, add `--diarize`)
 
 <h2 align="left", id="highlights">New🚨</h2>
 
@@ -247,6 +253,7 @@ Bug finding and pull requests are also highly appreciated to keep this project g
 
 <h2 align="left" id="contact">Contact/Support 📇</h2>
 
+
 Contact maxhbain@gmail.com for queries. WhisperX v4 development is underway with with siginificantly improved diarization. To support v4 and get early access, get in touch.
 
 <a href="https://www.buymeacoffee.com/maxhbain" target="_blank"><img src="https://cdn.buymeacoffee.com/buttons/default-orange.png" alt="Buy Me A Coffee" height="41" width="174"></a>
@@ -257,7 +264,9 @@ Contact maxhbain@gmail.com for queries. WhisperX v4 development is underway with
 This work, and my PhD, is supported by the [VGG (Visual Geometry Group)](https://www.robots.ox.ac.uk/~vgg/) and the University of Oxford.
 
 Of course, this is builds on [openAI's whisper](https://github.com/openai/whisper).
-And borrows important alignment code from [PyTorch tutorial on forced alignment](https://pytorch.org/tutorials/intermediate/forced_alignment_with_torchaudio_tutorial.html)
+Borrows important alignment code from [PyTorch tutorial on forced alignment](https://pytorch.org/tutorials/intermediate/forced_alignment_with_torchaudio_tutorial.html)
+And uses the wonderful pyannote VAD / Diarization https://github.com/pyannote/pyannote-audio
+
 
 Valuable VAD & Diarization Models from [pyannote audio][https://github.com/pyannote/pyannote-audio]
 

notebooks/whisperx.ipynb

@@ -0,0 +1,91 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "11fc5246",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/opt/conda/lib/python3.8/site-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: /opt/conda/lib/python3.8/site-packages/torchvision/image.so: undefined symbol: _ZNK3c1010TensorImpl36is_contiguous_nondefault_policy_implENS_12MemoryFormatE\n",
+      "  warn(f\"Failed to load image Python extension: {e}\")\n"
+     ]
+    },
+    {
+     "ename": "OutOfMemoryError",
+     "evalue": "CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 7.79 GiB total capacity; 5.76 GiB already allocated; 59.19 MiB free; 6.06 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mOutOfMemoryError\u001b[0m                          Traceback (most recent call last)",
+      "\u001b[0;32m/tmp/ipykernel_66/1447832577.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0;31m# transcribe with original whisper\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mwhisper\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"large\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      9\u001b[0m \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtranscribe\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0maudio_file\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     10\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/opt/conda/lib/python3.8/site-packages/whisper/__init__.py\u001b[0m in \u001b[0;36mload_model\u001b[0;34m(name, device, download_root, in_memory)\u001b[0m\n\u001b[1;32m    152\u001b[0m         \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_alignment_heads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0malignment_heads\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    153\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 154\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
+      "\u001b[0;32m/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36mto\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    987\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_floating_point\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_complex\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_blocking\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    988\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 989\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconvert\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    990\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    991\u001b[0m     def register_backward_hook(\n",
+      "\u001b[0;32m/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_apply\u001b[0;34m(self, fn)\u001b[0m\n\u001b[1;32m    639\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    640\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchildren\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 641\u001b[0;31m             \u001b[0mmodule\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    642\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    643\u001b[0m         \u001b[0;32mdef\u001b[0m \u001b[0mcompute_should_use_set_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtensor_applied\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_apply\u001b[0;34m(self, fn)\u001b[0m\n\u001b[1;32m    639\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    640\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchildren\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 641\u001b[0;31m             \u001b[0mmodule\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    642\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    643\u001b[0m         \u001b[0;32mdef\u001b[0m \u001b[0mcompute_should_use_set_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtensor_applied\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_apply\u001b[0;34m(self, fn)\u001b[0m\n\u001b[1;32m    639\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    640\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchildren\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 641\u001b[0;31m             \u001b[0mmodule\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    642\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    643\u001b[0m         \u001b[0;32mdef\u001b[0m \u001b[0mcompute_should_use_set_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtensor_applied\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_apply\u001b[0;34m(self, fn)\u001b[0m\n\u001b[1;32m    639\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    640\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchildren\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 641\u001b[0;31m             \u001b[0mmodule\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    642\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    643\u001b[0m         \u001b[0;32mdef\u001b[0m \u001b[0mcompute_should_use_set_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtensor_applied\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_apply\u001b[0;34m(self, fn)\u001b[0m\n\u001b[1;32m    639\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    640\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchildren\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 641\u001b[0;31m             \u001b[0mmodule\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    642\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    643\u001b[0m         \u001b[0;32mdef\u001b[0m \u001b[0mcompute_should_use_set_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtensor_applied\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_apply\u001b[0;34m(self, fn)\u001b[0m\n\u001b[1;32m    662\u001b[0m             \u001b[0;31m# `with torch.no_grad():`\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    663\u001b[0m             \u001b[0;32mwith\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mno_grad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 664\u001b[0;31m                 \u001b[0mparam_applied\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparam\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    665\u001b[0m             \u001b[0mshould_use_set_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcompute_should_use_set_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparam\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparam_applied\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    666\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mshould_use_set_data\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36mconvert\u001b[0;34m(t)\u001b[0m\n\u001b[1;32m    985\u001b[0m                 return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None,\n\u001b[1;32m    986\u001b[0m                             non_blocking, memory_format=convert_to_format)\n\u001b[0;32m--> 987\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_floating_point\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_complex\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_blocking\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    988\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    989\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_apply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconvert\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;31mOutOfMemoryError\u001b[0m: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 7.79 GiB total capacity; 5.76 GiB already allocated; 59.19 MiB free; 6.06 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF"
+     ]
+    }
+   ],
+   "source": [
+    "import whisperx\n",
+    "import whisper\n",
+    "\n",
+    "device = \"cuda\" \n",
+    "audio_file = \"audio.mp3\"\n",
+    "\n",
+    "# transcribe with original whisper\n",
+    "model = whisper.load_model(\"large\", device)\n",
+    "result = model.transcribe(audio_file)\n",
+    "\n",
+    "print(result[\"segments\"]) # before alignment\n",
+    "\n",
+    "# load alignment model and metadata\n",
+    "model_a, metadata = whisperx.load_align_model(language_code=result[\"language\"], device=device)\n",
+    "\n",
+    "# align whisper output\n",
+    "result_aligned = whisperx.align(result[\"segments\"], model_a, metadata, audio_file, device)\n",
+    "\n",
+    "print(result_aligned[\"segments\"]) # after alignment\n",
+    "print(result_aligned[\"word_segments\"]) # after alignment"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b63e6170",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

setup.py

@@ -7,7 +7,7 @@ setup(
     name="whisperx",
     py_modules=["whisperx"],
     version="3.1.0",
-    description="Time-Accurate Automatic Speech Recognition using Whisper.",
+    description="Time-Accurate Automatic Speech Recognition.",
     readme="README.md",
     python_requires=">=3.8",
     author="Max Bain",