update gradio demo

app.ipynb

@@ -69,35 +69,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
-   "id": "5f4d3586-a6b9-4d3e-b02a-9f25f5068dbe",
-   "metadata": {},
-   "outputs": [
-    {
-     "ename": "AttributeError",
-     "evalue": "module 'faster_whisper' has no attribute '__version__'",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[10], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mfaster_whisper\u001b[39;00m\n\u001b[0;32m----> 2\u001b[0m \u001b[43mfaster_whisper\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m__version__\u001b[49m\n",
-      "\u001b[0;31mAttributeError\u001b[0m: module 'faster_whisper' has no attribute '__version__'"
-     ]
-    }
-   ],
-   "source": [
-    "# import faster_whisper\n",
-    "# faster_whisper.__version__"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 5,
    "id": "de8e21b9-449a-4ae3-bd64-bba334075fdd",
    "metadata": {},
    "outputs": [],
    "source": [
-    "def t_asr(folder=\"vegam-whisper-medium-ml-fp16\", audio_file=\"00b38e80-80b8-4f70-babf-566e848879fc.webm\", compute_type=\"float16\", device=\"cpu\"):\n",
+    "def t_asr(folder=\"vegam-whisper-medium-ml-fp16\", audio_file=\"vegam-whisper-medium-ml-fp16/00b38e80-80b8-4f70-babf-566e848879fc.webm\", compute_type=\"float16\", device=\"cpu\"):\n",
     "    model = WhisperModel(folder, device=device, compute_type=compute_type)\n",
     "    \n",
     "    segments, info = model.transcribe(audio_file, beam_size=5)\n",
@@ -108,8 +85,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
-   "id": "87c58dd2-7d3d-4fb3-821c-cdac673fee0d",
+   "execution_count": 6,
+   "id": "25e1413f-8f80-4704-a94e-26b8d9581a6a",
    "metadata": {},
    "outputs": [
     {
@@ -117,68 +94,58 @@
      "output_type": "stream",
      "text": [
       "[0.00s -> 4.58s] പാലം കടുക്കുവോളം നാരായണ പാലം കടന്നാലോ കൂരായണ\n",
-      "CPU times: user 42.2 s, sys: 9.58 s, total: 51.8 s\n",
-      "Wall time: 13.5 s\n"
-     ]
-    }
-   ],
-   "source": [
-    "%%time\n",
-    "t_asr(compute_type=\"int8\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 28,
-   "id": "a5624cf5-b3b8-4ae3-aa82-ee19505bb42d",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Detected language 'ta' with probability 0.372757\n",
-      "[0.00s -> 4.74s] പാലം കടുക്കുവോളം നാരായണ പാലം കടന്നാലൊ കൂരായണ\n",
-      "CPU times: user 36.5 s, sys: 9.52 s, total: 46.1 s\n",
-      "Wall time: 12.3 s\n"
+      "CPU times: user 11.2 s, sys: 2.2 s, total: 13.4 s\n",
+      "Wall time: 6.54 s\n"
      ]
     }
    ],
    "source": [
     "%%time\n",
-    "t_asr(folder=\"vegam-whisper-medium-ml\", compute_type=\"int8\")"
+    "t_asr(compute_type=\"int8\", device=\"cuda\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
-   "id": "25e1413f-8f80-4704-a94e-26b8d9581a6a",
+   "execution_count": 7,
+   "id": "b255a0f0-4987-4f04-b63c-5ca0167917b6",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "[0.00s -> 4.58s] പാലം കടുക്കുവോളം നാരായണ പാലം കടന്നാലോ കൂരായണ\n",
-      "CPU times: user 9.39 s, sys: 792 ms, total: 10.2 s\n",
-      "Wall time: 4.51 s\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
-    "%%time\n",
-    "t_asr(compute_type=\"int8\", device=\"cuda\")"
+    "#|export \n",
+    "def transcribe_malayalam_speech(audio_file, compute_type=\"int8\", device=\"cpu\", folder=\"vegam-whisper-medium-ml-fp16\"):\n",
+    "    \n",
+    "    model = WhisperModel(folder, device=device, compute_type=compute_type)\n",
+    "    segments, info = model.transcribe(audio_file, beam_size=5)\n",
+    "\n",
+    "    lst = []\n",
+    "    for segment in segments:\n",
+    "        # print(\"[%.2fs -> %.2fs] %s\" % (segment.start, segment.end, segment.text))\n",
+    "        lst.append(segment.text)\n",
+    "\n",
+    "    return(\" \".join(lst))"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 8,
    "id": "48cd4ec3-512f-49d0-87ac-3ef989e25b80",
    "metadata": {},
    "outputs": [],
    "source": [
     "#|export\n",
-    "def transcribe_malayalam_speech(audio_file, compute_type=\"int8\", device=\"cpu\", folder=\"vegam-whisper-medium-ml-fp16\"):\n",
+    "def gr_transcribe_malayalam_speech(microphone, file_upload, compute_type=\"int8\", device=\"cpu\", folder=\"vegam-whisper-medium-ml-fp16\"):\n",
+    "    warn_output = \"\"\n",
+    "    if (microphone is not None) and (file_upload is not None):\n",
+    "        warn_output = (\n",
+    "            \"WARNING: You've uploaded an audio file and used the microphone. \"\n",
+    "            \"The recorded file from the microphone will be used and the uploaded audio will be discarded.\\n\"\n",
+    "        )\n",
+    "\n",
+    "    elif (microphone is None) and (file_upload is None):\n",
+    "        return \"ERROR: You have to either use the microphone or upload an audio file\"\n",
+    "\n",
+    "    audio_file = microphone if microphone is not None else file_upload\n",
+    "    \n",
     "    model = WhisperModel(folder, device=device, compute_type=compute_type)\n",
     "    segments, info = model.transcribe(audio_file, beam_size=5)\n",
     "\n",
@@ -192,7 +159,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 9,
    "id": "14fda29a-aee1-44b2-9269-048cc8b98ea8",
    "metadata": {},
    "outputs": [
@@ -200,8 +167,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "CPU times: user 43.1 s, sys: 12.3 s, total: 55.4 s\n",
-      "Wall time: 14.8 s\n"
+      "CPU times: user 40.6 s, sys: 9.76 s, total: 50.3 s\n",
+      "Wall time: 13.6 s\n"
      ]
     },
     {
@@ -210,14 +177,14 @@
        "'പാലം കടുക്കുവോളം നാരായണ പാലം കടന്നാലോ കൂരായണ'"
       ]
      },
-     "execution_count": 5,
+     "execution_count": 9,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
     "%%time\n",
-    "transcribe_malayalam_speech(audio_file=\"00b38e80-80b8-4f70-babf-566e848879fc.webm\")"
+    "transcribe_malayalam_speech(audio_file=\"vegam-whisper-medium-ml-fp16/00b38e80-80b8-4f70-babf-566e848879fc.webm\")"
    ]
   },
   {
@@ -230,14 +197,6 @@
     "## Haha, You are burning GPUs and wasting CO2"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "bf706a0a-c3a2-489c-a1fe-df4fbf700d9c",
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
   {
    "cell_type": "markdown",
    "id": "45fade75-e0b1-4c5d-90a3-ebd7345a4d16",
@@ -246,203 +205,6 @@
     "## Figure out Whisper  Demo by Huggingface"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": 36,
-   "id": "fa06f8a6-87b7-45af-b36b-fb5ebe362455",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e437727ccbcd40838a43a0c1bbb00143",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading (…)lve/main/config.json:   0%|          | 0.00/1.97k [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "2f654c303e24413cb73990bdd9d99907",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading pytorch_model.bin:   0%|          | 0.00/967M [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "16386d3b586d475fa021ea8d6f925161",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading (…)neration_config.json:   0%|          | 0.00/3.51k [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "bf5964b1ba024ce685a04127f21f78d0",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading (…)okenizer_config.json:   0%|          | 0.00/842 [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "038082a393084da998eed2085960e634",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading (…)olve/main/vocab.json:   0%|          | 0.00/1.04M [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "105ef799439d4c1ea0e3d2cbbfbcaf5d",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading (…)/main/tokenizer.json:   0%|          | 0.00/2.20M [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e3369330ed9a4a9f8208ba6f160210bf",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading (…)olve/main/merges.txt:   0%|          | 0.00/494k [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "4c3a9c73c84245b0b88e42980d65abdf",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading (…)main/normalizer.json:   0%|          | 0.00/52.7k [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "8d61551d78914036a2b6475a6d840663",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading (…)in/added_tokens.json:   0%|          | 0.00/2.08k [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "262213d3b6364b4e8648180c903c3008",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading (…)cial_tokens_map.json:   0%|          | 0.00/2.08k [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "258a3b7a9eb94dcdb8355c09c1b683b3",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Downloading (…)rocessor_config.json:   0%|          | 0.00/185k [00:00<?, ?B/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "import torch\n",
-    "from transformers import pipeline\n",
-    "from huggingface_hub import model_info\n",
-    "\n",
-    "MODEL_NAME = \"openai/whisper-small\" #this always needs to stay in line 8 :D sorry for the hackiness\n",
-    "lang = \"en\"\n",
-    "\n",
-    "device = 0 if torch.cuda.is_available() else \"cpu\"\n",
-    "pipe = pipeline(\n",
-    "    task=\"automatic-speech-recognition\",\n",
-    "    model=MODEL_NAME,\n",
-    "    chunk_length_s=30,\n",
-    "    device=device,\n",
-    ")\n",
-    "\n",
-    "pipe.model.config.forced_decoder_ids = pipe.tokenizer.get_decoder_prompt_ids(language=lang, task=\"transcribe\")\n",
-    "\n",
-    "def transcribe(microphone, file_upload):\n",
-    "    warn_output = \"\"\n",
-    "    if (microphone is not None) and (file_upload is not None):\n",
-    "        warn_output = (\n",
-    "            \"WARNING: You've uploaded an audio file and used the microphone. \"\n",
-    "            \"The recorded file from the microphone will be used and the uploaded audio will be discarded.\\n\"\n",
-    "        )\n",
-    "\n",
-    "    elif (microphone is None) and (file_upload is None):\n",
-    "        return \"ERROR: You have to either use the microphone or upload an audio file\"\n",
-    "\n",
-    "    file = microphone if microphone is not None else file_upload\n",
-    "\n",
-    "    text = pipe(file)[\"text\"]\n",
-    "\n",
-    "    return warn_output + text"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -461,7 +223,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
+   "execution_count": 10,
    "id": "9badfdcd-dd99-49ea-a318-eda88cddefb6",
    "metadata": {},
    "outputs": [
@@ -469,8 +231,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Running on local URL:  http://0.0.0.0:6007\n",
-      "Running on public URL: https://537af5b5b55ed185f5.gradio.live\n",
+      "Running on local URL:  http://0.0.0.0:6006\n",
+      "Running on public URL: https://9fa992d2ba37b0af49.gradio.live\n",
       "\n",
       "This share link expires in 72 hours. For free permanent hosting and GPU upgrades (NEW!), check out Spaces: https://huggingface.co/spaces\n"
      ]
@@ -478,7 +240,7 @@
     {
      "data": {
       "text/html": [
-       "<div><iframe src=\"https://537af5b5b55ed185f5.gradio.live\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+       "<div><iframe src=\"https://9fa992d2ba37b0af49.gradio.live\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
       ],
       "text/plain": [
        "<IPython.core.display.HTML object>"
@@ -491,7 +253,7 @@
      "data": {
       "text/plain": []
      },
-     "execution_count": 38,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -508,35 +270,80 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 41,
+   "execution_count": 20,
    "id": "81f3b241-8a6d-4ff0-bb70-d389d4d4e93a",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/opt/conda/lib/python3.10/site-packages/gradio/inputs.py:321: UserWarning: Usage of gradio.inputs is deprecated, and will not be supported in the future, please import your components from gradio.components\n",
+      "  warnings.warn(\n",
+      "/opt/conda/lib/python3.10/site-packages/gradio/inputs.py:324: UserWarning: `optional` parameter is deprecated, and it has no effect\n",
+      "  super().__init__(source=source, type=type, label=label, optional=optional)\n"
+     ]
+    }
+   ],
    "source": [
+    "#|export\n",
     "mf_transcribe = gr.Interface(\n",
-    "    fn=transcribe,\n",
+    "    fn=gr_transcribe_malayalam_speech,\n",
     "    inputs=[\n",
     "        gr.inputs.Audio(source=\"microphone\", type=\"filepath\", optional=True),\n",
     "        gr.inputs.Audio(source=\"upload\", type=\"filepath\", optional=True),\n",
     "    ],\n",
     "    outputs=\"text\",\n",
-    "    title=\"Whisper Demo: Transcribe Audio\",\n",
+    "    title=\"PALLAKKU (പല്ലക്ക്)\",\n",
     "    description=(\n",
-    "        \"Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the the fine-tuned\"\n",
-    "        f\" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe audio files\"\n",
-    "        \" of arbitrary length.\"\n",
+    "        \"Pallakku is a Malayalam speech to text demo leveraging the model-weights of [vegam-whisper-medium-ml](https://huggingface.co/kurianbenoy/vegam-whisper-medium-ml-fp16).\"\n",
     "    ),\n",
+    "    article=\"Please note that this demo now uses CPU only and in my testing for a 5 seconds audio file it can take upto 15 seconds for results to come. If you are interested to use a GPU based API instead, feel free to contact the author @ kurian.bkk@gmail.com\",\n",
     "    allow_flagging=\"never\",\n",
     ")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 21,
    "id": "b1e34fa5-8340-4329-a348-b641ca4db341",
    "metadata": {},
-   "outputs": [],
-   "source": []
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Running on local URL:  http://0.0.0.0:6010\n",
+      "Running on public URL: https://19b32861466405ac95.gradio.live\n",
+      "\n",
+      "This share link expires in 72 hours. For free permanent hosting and GPU upgrades (NEW!), check out Spaces: https://huggingface.co/spaces\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div><iframe src=\"https://19b32861466405ac95.gradio.live\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": []
+     },
+     "execution_count": 21,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#|export\n",
+    "mf_transcribe.launch(share=True)"
+   ]
   },
   {
    "cell_type": "markdown",
@@ -548,7 +355,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 22,
    "id": "7c3e753f-5051-4c3b-a5ab-fa65c7e7cae9",
    "metadata": {},
    "outputs": [
@@ -556,14 +363,15 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Overwriting requirements.txt\n"
+      "Writing requirements.txt\n"
      ]
     }
    ],
    "source": [
     "%%writefile requirements.txt\n",
     "gradio==3.31.0\n",
-    "faster-whisper==0.5.1"
+    "faster-whisper==0.5.1\n",
+    "torch"
    ]
   },
   {
@@ -576,7 +384,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 59,
+   "execution_count": 23,
    "id": "fba83810-1f0f-4777-b831-aabb4cfead39",
    "metadata": {},
    "outputs": [],