Upload folder using huggingface_hub

.github/workflows/update_space.yml

@@ -0,0 +1,28 @@
+name: Run Python script
+
+on:
+  push:
+    branches:
+      - main
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v2
+
+    - name: Set up Python
+      uses: actions/setup-python@v2
+      with:
+        python-version: '3.9'
+
+    - name: Install Gradio
+      run: python -m pip install gradio
+
+    - name: Log in to Hugging Face
+      run: python -c 'import huggingface_hub; huggingface_hub.login(token="${{ secrets.hf_token }}")'
+
+    - name: Deploy to Spaces
+      run: gradio deploy

.gradio/certificate.pem

@@ -0,0 +1,31 @@
+-----BEGIN CERTIFICATE-----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=
+-----END CERTIFICATE-----

.ipynb_checkpoints/ASPMIR-YorTTS-checkpoint.ipynb

@@ -0,0 +1,263 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 55,
+   "id": "23e98a8a-7128-4f35-ba1c-ff514ed462e0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Install Dependencies\n",
+    "#!pip3 install torch torchvision torchaudio\n",
+    "#!pip install transformers ipywidgets gradio --upgrade\n",
+    "#!pip install --upgrade gradio\n",
+    "#!pip install nltk\n",
+    "#!pip install jiwer\n",
+    "#!pip install sentencepiece\n",
+    "#!pip install sacremoses\n",
+    "#!pip install soundfile"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 56,
+   "id": "29275fa9-1b88-4e37-a278-7118bfca860a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\n",
+    "##translation_pipeline = pipeline('translation_en_to_fr')\n",
+    "##Evaluation Metric = BLEU score\n",
+    "##Exp1\n",
+    "#model_name = \"Davlan/byt5-base-eng-yor-mt\"\n",
+    "##Exp2\n",
+    "#model_name = \"Davlan/m2m100_418M-eng-yor-mt\" \n",
+    "##Exp3\n",
+    "#model_name = \"Davlan/mbart50-large-eng-yor-mt\"\n",
+    "##Exp4\n",
+    "#model_name = \"Davlan/mt5_base_eng_yor_mt\"\n",
+    "##Exp5\n",
+    "#model_name = \"omoekan/opus-tatoeba-eng-yor\"\n",
+    "##Exp6\n",
+    "#model_name = \"masakhane/afrimt5_en_yor_news\"\n",
+    "##Exp7\n",
+    "#model_name = \"masakhane/afrimbart_en_yor_news\"\n",
+    "##Exp8\n",
+    "#model_name = \"masakhane/afribyt5_en_yor_news\"\n",
+    "##Exp9\n",
+    "#model_name = \"masakhane/byt5_en_yor_news\"\n",
+    "##Exp10\n",
+    "#model_name = \"masakhane/mt5_en_yor_news\"\n",
+    "#translation_pipeline = pipeline('translation_en_to_yo', model = model_name, max_length=50)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 57,
+   "id": "1ea4a2eb-6cbf-497a-a080-2db3dd64be36",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#results = translation_pipeline('My Name is Ayo, I love books')\n",
+    "#results[0]['translation_text']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 58,
+   "id": "f92487b5-158a-47ef-ab12-a361ea8d0a48",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#results = translation_pipeline('The wages of sin is death')\n",
+    "#results[0]['translation_text']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 59,
+   "id": "69d64db9-b083-46ae-80ce-9616ba99183d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from transformers import pipeline\n",
+    "import nltk\n",
+    "import jiwer\n",
+    "from nltk.translate.bleu_score import corpus_bleu\n",
+    "from transformers import VitsModel, AutoTokenizer\n",
+    "import torch\n",
+    "import soundfile as sf\n",
+    "\n",
+    "\n",
+    "WerScore = 0\n",
+    "bleuScore = 0\n",
+    "def translate_transformers(modelName, sourceLangText):\n",
+    "    #results = translation_pipeline(input_text)\n",
+    "    translation_pipeline = pipeline('translation_en_to_yo', model = modelName, max_length=500)\n",
+    "    translated_text = translation_pipeline(sourceLangText) #translator(text)[0][\"translation_text\"]\n",
+    "    translated_text_target = translated_text[0]['translation_text']\n",
+    "    hypothesis_translations = \"My name is Joy, I love reading\"\n",
+    "   \n",
+    "    #TTS for the translated_text_target\n",
+    "    #TTS Exp1\n",
+    "    ttsModel = VitsModel.from_pretrained(\"facebook/mms-tts-yor\")\n",
+    "    tokenizer = AutoTokenizer.from_pretrained(\"facebook/mms-tts-yor\")\n",
+    "    ttsInputs = tokenizer(translated_text_target, return_tensors=\"pt\")\n",
+    "   \n",
+    "    with torch.no_grad():\n",
+    "        ttsOutput = ttsModel(**ttsInputs).waveform\n",
+    "    #onvert the tensor to a numpy array\n",
+    "    ttsWaveform = ttsOutput.numpy()[0]    \n",
+    "    #Save the waveform to an audio file\n",
+    "    #sf.write('output.wav', waveform, 22050)\n",
+    "    sf.write('ttsOutput.wav', ttsWaveform, 16000)\n",
+    "    \n",
+    "    #Calculate WerScore\n",
+    "    WerScore = jiwer.wer(translated_text_target, hypothesis_translations)\n",
+    "    #bleuScore = corpus_bleu(translated_text_target,hypothesis_translations)\n",
+    "    \n",
+    "    return translated_text_target,WerScore,'ttsOutput.wav'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "id": "5d9ed5a2-0d28-4078-923d-c8c27196292a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#text1 = \"Oruko mi ni Ayo, mo feran iwe kika gan\"\n",
+    "#text2 = \"Agbaninímọ̀ràn kan lórí ọ̀ràn radiation and Clinical Oncologist, tórúkọ rẹ̀ ń jẹ́ Temitope Olatunji-Agunbiade ti kìlọ̀ fáwọn obìnrin pé kí wọ́n má ṣe lo oògùn máàjóyúndúró tàbí kí wọ́n lo oògùn máàjóyúndúró, ó sọ pé ìwádìí ti fi hàn pé lílò tí wọ́n ń lò ó ń mú kí ewu àrùn jẹjẹrẹ ọmú pọ̀ sí i.\"\n",
+    "\n",
+    "#with torch.no_grad():\n",
+    "    #output = ttsModel(**inputs).waveform"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "id": "54138308-b423-4e7c-9469-2002bfeb7918",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#from IPython.display import Audio\n",
+    "#Audio(output, rate=ttsModel.config.sampling_rate)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 62,
+   "id": "bbf259d6-922d-4f5c-9af1-cbd57158a814",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Running on local URL:  http://127.0.0.1:7879\n",
+      "Running on public URL: https://ccee705195aed67b23.gradio.live\n",
+      "\n",
+      "This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div><iframe src=\"https://ccee705195aed67b23.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": 62,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#Gradio Function and Interface\n",
+    "import gradio as gr\n",
+    "from IPython.display import Audio\n",
+    "interface = gr.Interface(\n",
+    "    fn=translate_transformers,\n",
+    "    inputs=[\n",
+    "        gr.Dropdown([\"Davlan/byt5-base-eng-yor-mt\", #Exp1\n",
+    "                     \"Davlan/m2m100_418M-eng-yor-mt\", #Exp2\n",
+    "                     \"Davlan/mbart50-large-eng-yor-mt\", #Exp3\n",
+    "                     \"Davlan/mt5_base_eng_yor_mt\", #Exp4\n",
+    "                     \"omoekan/opus-tatoeba-eng-yor\", #Exp5\n",
+    "                     \"masakhane/afrimt5_en_yor_news\", #Exp6\n",
+    "                     \"masakhane/afrimbart_en_yor_news\", #Exp7\n",
+    "                     \"masakhane/afribyt5_en_yor_news\", #Exp8\n",
+    "                     \"masakhane/byt5_en_yor_news\", #Exp9\n",
+    "                     \"masakhane/mt5_en_yor_news\", #Exp10\n",
+    "                     \"masakhane/mbart50_en_yor_news\", #Exp11\n",
+    "                     \"masakhane/m2m100_418M_en_yor_news\", #Exp12\n",
+    "                     \"masakhane/m2m100_418M_en_yor_rel_news\", #Exp13\n",
+    "                     \"masakhane/m2m100_418M_en_yor_rel_news_ft\", #Exp14\n",
+    "                     \"masakhane/m2m100_418M_en_yor_rel\", #Exp15\n",
+    "                     #\"facebook/nllb-200-distilled-600M\", #Exp16\n",
+    "                     #\"facebook/nllb-200-3.3B\", #Exp17\n",
+    "                     #\"facebook/nllb-200-1.3B\", #Exp18\n",
+    "                     #\"facebook/nllb-200-distilled-1.3B\",  #Exp19\n",
+    "                     #\"keithhon/nllb-200-3.3B\" #Exp20\n",
+    "                     #\"CohereForAI/aya-101\" #Exp16\n",
+    "                     ], \n",
+    "                     label=\"Select Finetuned Eng2Yor Translation Model\"),\n",
+    "        gr.Textbox(lines=2, placeholder=\"Enter English Text Here...\", label=\"English Text\")  \n",
+    "    ],\n",
+    "    #outputs = \"text\",\n",
+    "    #outputs=outputs=[\"text\", \"text\"],#\"text\"\n",
+    "    #outputs= gr.Textbox(value=\"text\", label=\"Translated Text\"),\n",
+    "    outputs=[\n",
+    "        gr.Textbox(value=\"text\", label=\"Translated Yoruba Text\"),\n",
+    "        #gr.Textbox(value=\"text\", label=translated_text_actual),\n",
+    "        gr.Textbox(value=\"number\", label=\"WER(Word Error Rate) Score - The Lower the Better\"),\n",
+    "        #gr.Textbox(value=\"number\", label=\"Bleu Score\")\n",
+    "        gr.Audio(type=\"filepath\", label=\"Click to Generate Yoruba Text2Speech\")\n",
+    "    ],\n",
+    "    title=\"ASPMIR NEURAL MACHINE TRANSLATION(NMT) TESTBED FOR LOW RESOURCED AFRICAN LANGUAGES\",\n",
+    "    description=\"{This Tool Allows Developers and Researchers to Carry Out Experiments on Low Resourced African Languages with State-of-the-Art NMT Finetuned Models.}\"\n",
+    ")\n",
+    "\n",
+    "interface.launch(share=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c3baee0f-fd85-4209-9d54-14451abd372a",
+   "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": "3.10.10"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

.ipynb_checkpoints/requirements-checkpoint.txt

@@ -0,0 +1,283 @@
+absl-py==1.4.0
+accelerate==1.2.0
+aiofiles==23.2.1
+aiohttp==3.9.0
+aiohttp-cors==0.7.0
+aiosignal==1.3.1
+alembic==1.12.1
+altair==5.2.0
+annotated-types==0.6.0
+anyio==4.0.0
+argon2-cffi==23.1.0
+argon2-cffi-bindings==21.2.0
+array-record==0.5.0
+arrow==1.3.0
+asttokens==2.4.1
+astunparse==1.6.3
+async-generator==1.10
+async-lru==2.0.4
+async-timeout==4.0.3
+attrs==23.1.0
+audioread==3.0.1
+Babel==2.13.1
+beautifulsoup4==4.12.2
+bleach==6.1.0
+blessed==1.20.0
+boltons @ file:///home/conda/feedstock_root/build_artifacts/boltons_1677499911949/work
+brotlipy @ file:///home/conda/feedstock_root/build_artifacts/brotlipy_1666764671472/work
+cachetools==5.3.2
+certifi @ file:///home/conda/feedstock_root/build_artifacts/certifi_1707022139797/work/certifi
+certipy==0.1.3
+cffi @ file:///home/conda/feedstock_root/build_artifacts/cffi_1671179353105/work
+charset-normalizer @ file:///home/conda/feedstock_root/build_artifacts/charset-normalizer_1678108872112/work
+click==8.1.7
+colorama @ file:///home/conda/feedstock_root/build_artifacts/colorama_1666700638685/work
+colorful==0.5.5
+comm==0.2.0
+conda==23.3.1
+conda-package-handling @ file:///home/conda/feedstock_root/build_artifacts/conda-package-handling_1669907009957/work
+conda_package_streaming @ file:///home/conda/feedstock_root/build_artifacts/conda-package-streaming_1669733752472/work
+contourpy==1.2.0
+cryptography @ file:///home/conda/feedstock_root/build_artifacts/cryptography-split_1679811212387/work
+cycler==0.12.1
+debugpy==1.8.0
+decorator==5.1.1
+defusedxml==0.7.1
+distlib==0.3.7
+dm-tree==0.1.8
+et-xmlfile==1.1.0
+etils==1.6.0
+exceptiongroup==1.1.3
+executing==2.0.1
+fastapi==0.115.6
+fastjsonschema==2.19.0
+ffmpy==0.3.1
+filelock @ file:///home/conda/feedstock_root/build_artifacts/filelock_1698714947081/work
+flatbuffers==23.5.26
+fonttools==4.44.3
+fqdn==1.5.1
+frozenlist==1.4.0
+fsspec==2024.2.0
+gast==0.5.4
+GDAL @ file:///home/conda/feedstock_root/build_artifacts/gdal-split_1680712150998/work/build/swig/python
+gmpy2 @ file:///home/conda/feedstock_root/build_artifacts/gmpy2_1666808654411/work
+google-api-core==2.14.0
+google-auth==2.23.4
+google-auth-oauthlib==1.1.0
+google-pasta==0.2.0
+googleapis-common-protos==1.61.0
+gpustat==1.1.1
+gradio==5.8.0
+gradio_client==1.5.1
+greenlet==3.0.1
+grpcio==1.59.3
+gTTS==2.5.1
+h11==0.14.0
+h5py==3.10.0
+httpcore==1.0.2
+httpx==0.26.0
+huggingface-hub==0.26.3
+idna @ file:///home/conda/feedstock_root/build_artifacts/idna_1663625384323/work
+importlib-resources==6.1.1
+ipykernel==6.26.0
+ipython==8.17.2
+ipywidgets==8.1.2
+isoduration==20.11.0
+jedi==0.19.1
+Jinja2 @ file:///home/conda/feedstock_root/build_artifacts/jinja2_1654302431367/work
+jiwer==3.0.3
+joblib @ file:///home/conda/feedstock_root/build_artifacts/joblib_1691577114857/work
+json5==0.9.14
+jsonpatch @ file:///home/conda/feedstock_root/build_artifacts/jsonpatch_1695536281965/work
+jsonpointer @ file:///home/conda/feedstock_root/build_artifacts/jsonpointer_1695397238043/work
+jsonschema==4.20.0
+jsonschema-specifications==2023.11.1
+jupyter-events==0.9.0
+jupyter-lsp==2.2.0
+jupyter-resource-usage==1.0.1
+jupyter-telemetry==0.1.0
+jupyter_client==8.6.0
+jupyter_core==5.5.0
+jupyter_server==2.10.1
+jupyter_server_terminals==0.4.4
+jupyterhub==4.0.2
+jupyterlab==4.0.9
+jupyterlab-pygments==0.2.2
+jupyterlab_server==2.25.2
+jupyterlab_widgets==3.0.10
+keras==2.15.0
+kiwisolver==1.4.5
+lazy_loader==0.4
+libclang==16.0.6
+libmambapy @ file:///home/conda/feedstock_root/build_artifacts/mamba-split_1680002410624/work/libmambapy
+librosa==0.10.2.post1
+llvmlite==0.43.0
+lxml==5.1.0
+Mako==1.3.0
+mamba @ file:///home/conda/feedstock_root/build_artifacts/mamba-split_1680002410624/work/mamba
+Markdown==3.5.1
+markdown-it-py==3.0.0
+MarkupSafe @ file:///home/conda/feedstock_root/build_artifacts/markupsafe_1695367434228/work
+matlab-kernel==0.17.1
+matplotlib==3.8.2
+matplotlib-inline==0.1.6
+mdurl==0.1.2
+metakernel==0.30.2
+mistune==3.0.2
+ml-dtypes==0.2.0
+mpmath @ file:///home/conda/feedstock_root/build_artifacts/mpmath_1678228039184/work
+msgpack==1.0.7
+multidict==6.0.4
+nbclient==0.9.0
+nbconvert==7.11.0
+nbformat==5.9.2
+nbgitpuller==1.2.0
+nest-asyncio==1.5.8
+networkx @ file:///home/conda/feedstock_root/build_artifacts/networkx_1698504735452/work
+nltk==3.8.1
+notebook==7.0.6
+notebook_shim==0.2.3
+numba==0.60.0
+numpy @ file:///home/conda/feedstock_root/build_artifacts/numpy_1695290862901/work/dist/numpy-1.26.0-cp310-cp310-linux_x86_64.whl#sha256=44509c98ccedaff13cf312f80a8e392a35d5f649bdf63f1b7e705fdfdc621c6d
+nvidia-cublas-cu12==12.1.3.1
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-cuda-nvcc-cu12==12.2.140
+nvidia-cuda-nvrtc-cu12==12.1.105
+nvidia-cuda-runtime-cu12==12.1.105
+nvidia-cudnn-cu12==8.9.2.26
+nvidia-cufft-cu12==11.0.2.54
+nvidia-curand-cu12==10.3.2.106
+nvidia-cusolver-cu12==11.4.5.107
+nvidia-cusparse-cu12==12.1.0.106
+nvidia-ml-py==12.535.133
+nvidia-nccl-cu12==2.20.5
+nvidia-nvjitlink-cu12==12.2.140
+nvidia-nvtx-cu12==12.1.105
+oauthlib==3.2.2
+opencensus==0.11.3
+opencensus-context==0.1.3
+opencv-python==4.8.1.78
+openpyxl==3.1.3
+opt-einsum==3.3.0
+orjson==3.9.13
+overrides==7.4.0
+packaging @ file:///home/conda/feedstock_root/build_artifacts/packaging_1696202382185/work
+pamela==1.1.0
+pandas==2.1.3
+pandocfilters==1.5.0
+parso==0.8.3
+pexpect==4.8.0
+Pillow==10.1.0
+platformdirs==3.11.0
+pluggy @ file:///home/conda/feedstock_root/build_artifacts/pluggy_1667232663820/work
+pooch==1.8.2
+portalocker==2.8.2
+prometheus-client==0.18.0
+promise==2.3
+prompt-toolkit==3.0.41
+protobuf==3.20.3
+psutil==5.9.6
+ptyprocess==0.7.0
+pure-eval==0.2.2
+py-spy==0.3.14
+pyasn1==0.5.0
+pyasn1-modules==0.3.0
+pycosat @ file:///home/conda/feedstock_root/build_artifacts/pycosat_1666836542287/work
+pycparser @ file:///home/conda/feedstock_root/build_artifacts/pycparser_1636257122734/work
+pydantic==2.6.1
+pydantic_core==2.16.2
+pydub==0.25.1
+Pygments==2.17.0
+pyOpenSSL @ file:///home/conda/feedstock_root/build_artifacts/pyopenssl_1680037383858/work
+pyparsing==3.1.1
+PySocks @ file:///home/conda/feedstock_root/build_artifacts/pysocks_1661604839144/work
+python-dateutil==2.8.2
+python-json-logger==2.0.7
+python-multipart==0.0.19
+pytz==2023.3.post1
+PyYAML==6.0.1
+pyzmq==25.1.1
+rapidfuzz==3.6.1
+ray==2.8.0
+referencing==0.31.0
+regex==2023.12.25
+requests==2.31.0
+requests-oauthlib==1.3.1
+rfc3339-validator==0.1.4
+rfc3986-validator==0.1.1
+rich==13.7.0
+rpds-py==0.13.0
+rsa==4.9
+ruamel.yaml @ file:///home/conda/feedstock_root/build_artifacts/ruamel.yaml_1678272977710/work
+ruamel.yaml.clib @ file:///home/conda/feedstock_root/build_artifacts/ruamel.yaml.clib_1670412719074/work
+ruff==0.2.2
+sacrebleu==2.4.0
+sacremoses==0.1.1
+safehttpx==0.1.6
+safetensors==0.4.5
+scapy==2.5.0
+scikit-learn @ file:///home/conda/feedstock_root/build_artifacts/scikit-learn_1698224870717/work
+scipy @ file:///home/conda/feedstock_root/build_artifacts/scipy-split_1696467628975/work/dist/scipy-1.11.3-cp310-cp310-linux_x86_64.whl#sha256=1a1f6f0b1d49eca3673dad934971d667767456b5f7effcaf9f07cdd3ba377c95
+seaborn==0.13.2
+semantic-version==2.10.0
+Send2Trash==1.8.2
+sentencepiece==0.2.0
+shellingham==1.5.4
+six==1.16.0
+smart-open==6.4.0
+sniffio==1.3.0
+soundfile==0.12.1
+soupsieve==2.5
+soxr==0.5.0.post1
+spicy==0.16.0
+split-folders==0.5.1
+SQLAlchemy==2.0.23
+stack-data==0.6.3
+starlette==0.41.3
+sympy @ file:///home/conda/feedstock_root/build_artifacts/sympy_1684180540116/work
+tabulate==0.9.0
+tensorboard==2.15.1
+tensorboard-data-server==0.7.2
+tensorflow==2.15.0.post1
+tensorflow-estimator==2.15.0
+tensorflow-io-gcs-filesystem==0.34.0
+tensorflow-metadata==1.14.0
+termcolor==2.4.0
+terminado==0.18.0
+tfds-nightly==4.9.3.dev202312070044
+there==0.0.12
+threadpoolctl @ file:///home/conda/feedstock_root/build_artifacts/threadpoolctl_1689261241048/work
+tinycss2==1.2.1
+tokenizers==0.21.0
+toml==0.10.2
+tomli==2.0.1
+tomlkit==0.12.0
+toolz @ file:///home/conda/feedstock_root/build_artifacts/toolz_1657485559105/work
+torch==2.3.1
+torchaudio==2.1.1+cu118
+torchvision==0.16.1+cu118
+tornado==6.3.3
+tqdm @ file:///home/conda/feedstock_root/build_artifacts/tqdm_1677948868469/work
+traitlets==5.13.0
+transformers==4.47.0
+triton==2.3.1
+typer==0.12.3
+types-python-dateutil==2.8.19.14
+typing_extensions @ file:///home/conda/feedstock_root/build_artifacts/typing_extensions_1695040754690/work
+tzdata==2023.3
+uri-template==1.3.0
+urllib3==2.2.1
+uvicorn==0.27.0.post1
+virtualenv==20.21.0
+wcwidth==0.2.10
+webcolors==1.13
+webencodings==0.5.1
+websocket-client==1.6.4
+websockets==11.0.3
+Werkzeug==3.0.1
+widgetsnbextension==4.0.10
+wrapt==1.14.1
+wurlitzer==3.1.1
+yarl==1.9.2
+zipp==3.17.0
+zstandard==0.19.0

ASPMIR-YorTTS.ipynb

@@ -0,0 +1,402 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "23e98a8a-7128-4f35-ba1c-ff514ed462e0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Install All the Required Dependencies\n",
+    "#!pip3 install torch torchvision torchaudio\n",
+    "#!pip install transformers ipywidgets gradio --upgrade\n",
+    "#!pip install --upgrade transformers accelerate\n",
+    "#!pip install --upgrade gradio\n",
+    "#!pip install nltk\n",
+    "#!pip install jiwer\n",
+    "#!pip install sentencepiece\n",
+    "#!pip install sacremoses\n",
+    "#!pip install soundfile\n",
+    "#!pip install librosa numpy jiwer nltk\n",
+    "#!pip install --upgrade pip \n",
+    "#!pip install huggingface_hub"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "0d2a7d3a-8c2c-4134-a79f-a3b7b1747874",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2024-12-20 20:13:51.723870: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
+      "2024-12-20 20:13:51.767697: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
+      "2024-12-20 20:13:51.767728: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
+      "2024-12-20 20:13:51.768839: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
+      "2024-12-20 20:13:51.775965: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
+      "To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
+      "2024-12-20 20:13:52.795860: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
+     ]
+    }
+   ],
+   "source": [
+    "#Import Required Libraries\n",
+    "from transformers import pipeline\n",
+    "from jiwer import wer\n",
+    "from transformers import VitsModel, AutoTokenizer, set_seed\n",
+    "import torch\n",
+    "import soundfile as sf\n",
+    "import librosa\n",
+    "from scipy.spatial.distance import euclidean\n",
+    "import numpy as np\n",
+    "import string\n",
+    "import os\n",
+    "from nltk.translate.bleu_score import sentence_bleu, SmoothingFunction\n",
+    "from nltk.translate.meteor_score import meteor_score\n",
+    "import string\n",
+    "import numpy as np\n",
+    "import librosa\n",
+    "from scipy.spatial.distance import euclidean\n",
+    "import string\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "e2bafb31-ecf6-44e4-b25a-24abfa75bed1",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "['/home/jupyter-prof-adetiba/nltk_data', '/opt/tljh/user/nltk_data', '/opt/tljh/user/share/nltk_data', '/opt/tljh/user/lib/nltk_data', '/usr/share/nltk_data', '/usr/local/share/nltk_data', '/usr/lib/nltk_data', '/usr/local/lib/nltk_data']\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[nltk_data] Downloading package wordnet to /home/jupyter-prof-\n",
+      "[nltk_data]     adetiba/nltk_data...\n",
+      "[nltk_data]   Package wordnet is already up-to-date!\n",
+      "[nltk_data] Downloading package omw-1.4 to /home/jupyter-prof-\n",
+      "[nltk_data]     adetiba/nltk_data...\n",
+      "[nltk_data]   Package omw-1.4 is already up-to-date!\n"
+     ]
+    }
+   ],
+   "source": [
+    "import nltk\n",
+    "nltk.download('wordnet')\n",
+    "nltk.download('omw-1.4')  # Optional if using WordNet's multilingual features\n",
+    "import nltk\n",
+    "print(nltk.data.path)\n",
+    "import nltk\n",
+    "nltk.data.path.append('./nltk_data')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "10ceb8b4-fe4e-4a97-ac34-dce6a890455a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Define all Utility Functions\n",
+    "# Function to compute BLEU score\n",
+    "def compute_bleu(reference_text, predicted_text):\n",
+    "    \"\"\"\n",
+    "    Computes the BLEU score for a single translation.\n",
+    "    :param reference_text: The ground truth text (in Yoruba).\n",
+    "    :param predicted_text: The machine-generated translation text (in Yoruba).\n",
+    "    :return: BLEU score (float).\n",
+    "    \"\"\"\n",
+    "    print(\"The Reference Text = \", reference_text)\n",
+    "    print(\"The Predicted Text = \",predicted_text)\n",
+    "    # Tokenize the reference and predicted texts\n",
+    "    reference_tokens = [reference_text.split()]  # Reference should be wrapped in a list\n",
+    "    predicted_tokens = predicted_text.split()\n",
+    "\n",
+    "    # Add smoothing to handle cases with few n-gram matches\n",
+    "    smoothing_function = SmoothingFunction().method1\n",
+    "\n",
+    "    # Compute BLEU score\n",
+    "    bleu_score = sentence_bleu(reference_tokens, predicted_tokens, smoothing_function=smoothing_function)\n",
+    "    #print(\"The Computed bleu_score in the Compute_Blue Fn = \",bleu_score)\n",
+    "    return round(bleu_score,2)\n",
+    "# Function to compute Word Error Rate (WER)\n",
+    "def compute_wer(reference_text, predicted_text):\n",
+    "    \"\"\"\n",
+    "    Computes the Word Error Rate (WER) for a single translation.\n",
+    "    :param reference_text: The ground truth text (in Yoruba).\n",
+    "    :param predicted_text: The machine-generated translation text (in Yoruba).\n",
+    "    :return: WER score (float).\n",
+    "    \"\"\"\n",
+    "    # Normalize text: lowercase and remove punctuation\n",
+    "    reference_text = reference_text.lower().translate(str.maketrans('', '', string.punctuation))\n",
+    "    predicted_text = predicted_text.lower().translate(str.maketrans('', '', string.punctuation))\n",
+    "\n",
+    "    # Compute WER\n",
+    "    wer_score = wer(reference_text, predicted_text)\n",
+    "\n",
+    "    return round(wer_score,2)\n",
+    "\n",
+    "# Function to compute METEOR score\n",
+    "def compute_meteor(reference_text, predicted_text):\n",
+    "    \"\"\"\n",
+    "    Computes the METEOR score for a single translation.\n",
+    "    :param reference_text: The ground truth text (in Yoruba).\n",
+    "    :param predicted_text: The machine-generated translation text (in Yoruba).\n",
+    "    :return: METEOR score (float).\n",
+    "    \"\"\"\n",
+    "    # Normalize text: lowercase and remove punctuation\n",
+    "    reference_text = reference_text.lower().translate(str.maketrans('', '', string.punctuation))\n",
+    "    predicted_text = predicted_text.lower().translate(str.maketrans('', '', string.punctuation))\n",
+    "\n",
+    "    # Tokenize text into lists of words\n",
+    "    reference_tokens = reference_text.split()\n",
+    "    predicted_tokens = predicted_text.split()\n",
+    "\n",
+    "    # Compute METEOR score\n",
+    "    meteor = meteor_score([reference_tokens], predicted_tokens)\n",
+    "    \n",
+    "    return round(meteor,2)\n",
+    "\n",
+    "# Function to compute Mel Cepstral Distance (MCD)\n",
+    "def compute_mcd(ground_truth_audio_path, predicted_audio_path):\n",
+    "    \"\"\"\n",
+    "    Computes the Mel Cepstral Distance (MCD) between two audio files.\n",
+    "    :param ground_truth_audio_path: Path to the ground truth audio file.\n",
+    "    :param predicted_audio_path: Path to the predicted audio file.\n",
+    "    :return: MCD score (float).\n",
+    "    \"\"\"\n",
+    "    # Load audio files\n",
+    "    y_true, sr_true = librosa.load(ground_truth_audio_path, sr=16000)\n",
+    "    y_pred, sr_pred = librosa.load(predicted_audio_path, sr=16000)\n",
+    "\n",
+    "    # Ensure the sampling rates match\n",
+    "    assert sr_true == sr_pred, \"Sampling rates do not match between audio files.\"\n",
+    "\n",
+    "    # Compute MFCCs\n",
+    "    mfcc_true = librosa.feature.mfcc(y=y_true, sr=sr_true, n_mfcc=13).T\n",
+    "    mfcc_pred = librosa.feature.mfcc(y=y_pred, sr=sr_pred, n_mfcc=13).T\n",
+    "\n",
+    "    # Align the MFCC frames\n",
+    "    min_frames = min(len(mfcc_true), len(mfcc_pred))\n",
+    "    mfcc_true = mfcc_true[:min_frames]\n",
+    "    mfcc_pred = mfcc_pred[:min_frames]\n",
+    "\n",
+    "    # Compute the Euclidean distance for each frame and average\n",
+    "    mcd = 0.0\n",
+    "    for i in range(min_frames):\n",
+    "        mcd += euclidean(mfcc_true[i], mfcc_pred[i])\n",
+    "    mcd = (10.0 / np.log(10)) * (mcd / min_frames)\n",
+    "\n",
+    "    return round(mcd,2)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "69d64db9-b083-46ae-80ce-9616ba99183d",
+   "metadata": {
+    "editable": true,
+    "slideshow": {
+     "slide_type": ""
+    },
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "#Define Translation and Synthesis Function\n",
+    "def translate_transformers(modelName, sourceLangText):\n",
+    "    #results = translation_pipeline(input_text)\n",
+    "    translation_pipeline = pipeline('translation_en_to_yo', model = modelName, max_length=500)\n",
+    "    translated_text = translation_pipeline(sourceLangText) #translator(text)[0][\"translation_text\"]\n",
+    "    translated_text_target = translated_text[0]['translation_text']\n",
+    "    #reference_translations = \"awon apositeli, awon woli, awon ajinrere ati awon oluso agutan ati awon oluko.\" #'recorder_2024-01-13_11-24-41_453538.wav'#\"My name is Joy, I love reading\"\n",
+    "   \n",
+    "    #TTS for the translated_text_target\n",
+    "    #TTS Exp1\n",
+    "    ttsModel = VitsModel.from_pretrained(\"facebook/mms-tts-yor\")\n",
+    "    tokenizer = AutoTokenizer.from_pretrained(\"facebook/mms-tts-yor\")\n",
+    "    ttsInputs = tokenizer(translated_text_target, return_tensors=\"pt\")\n",
+    "    set_seed(555)  # make deterministic\n",
+    "    with torch.no_grad():\n",
+    "        ttsOutput = ttsModel(**ttsInputs).waveform\n",
+    "    #Convert the tensor to a numpy array\n",
+    "    ttsWaveform = ttsOutput.numpy()[0]    \n",
+    "    #Save the waveform to an audio file\n",
+    "    #sf.write('output.wav', waveform, 22050)\n",
+    "    sf.write('ttsOutput.wav', ttsWaveform, 16000)\n",
+    "    \n",
+    "    # Sample ground truth and predicted text2text translations for Clinical Text\n",
+    "    #ground_truth_text = \"Àrùn jẹjẹrẹ ọmú jẹ́ ọ̀kan pàtàkì lára ohun tó ń ṣàkóbá fún ìlera gbogbo ènìyàn ní Nàìjíríà, ó sì jẹ́ ọ̀kan pàtàkì lára ohun tó ń fa ikú àwọn obìnrin tí àrùn jẹjẹrẹ ń pa lórílẹ̀-èdè náà.\"\n",
+    "    #predicted_text = translated_text_target #\"<extra_id_0> breast cancer is a\"\n",
+    "\n",
+    "    # Sample ground truth and predicted text2text translations for News Text\n",
+    "    #ground_truth_text = \"Wọ́n ní ìgbà àkọ́kọ́ nìyí tí irú ìwà ipá bẹ́ẹ̀ máa wáyé ní ìpínlẹ̀ Ondo.\"\n",
+    "    #predicted_text = translated_text_target #\"<extra_id_0> breast cancer is a\"\n",
+    "\n",
+    "    # Sample ground truth and predicted text2text translations for Religion Text\n",
+    "    ground_truth_text = \"Àwọn aposteli, àwọn wòlíì, àwọn ajíhìnrere, àwọn olùṣọ́-àgùntàn àti àwọn olùkọ́.\"\n",
+    "    predicted_text = translated_text_target #\"<extra_id_0> breast cancer is a\"\n",
+    "    \n",
+    "    #Compute bleu_score\n",
+    "    bleu_score = compute_bleu(ground_truth_text, predicted_text)\n",
+    "    print(f\"Bleu Score (BLEU): {bleu_score:.2f}\")\n",
+    "    \n",
+    "    #Compute WER\n",
+    "    wer_score = compute_wer(ground_truth_text, predicted_text)\n",
+    "    print(f\"Word Error Rate (WER): {wer_score:.2f}\")\n",
+    "\n",
+    "    #Compute METEOR\n",
+    "    meteor = compute_meteor(ground_truth_text, predicted_text)\n",
+    "    print(f\"METEOR Score: {meteor:.2f}\")\n",
+    "\n",
+    "    # Paths to sample audio files for MCD computation in current directory\n",
+    "    ground_truth_audio = os.path.join(os.getcwd(), \"gt_ttsOutput.wav\")\n",
+    "    predicted_audio = os.path.join(os.getcwd(), \"ttsOutput.wav\")\n",
+    "\n",
+    "    # Compute Mel Cepstral Distance (MCD)\n",
+    "    try:\n",
+    "        mcd = compute_mcd(ground_truth_audio, predicted_audio)\n",
+    "        print(f\"Mel Cepstral Distance (MCD): {mcd:.2f}\")\n",
+    "    except Exception as e:\n",
+    "        print(f\"Error computing MCD: {e}\")\n",
+    "    \n",
+    "    return translated_text_target,bleu_score,wer_score,meteor,mcd,'ttsOutput.wav'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "bbf259d6-922d-4f5c-9af1-cbd57158a814",
+   "metadata": {
+    "editable": true,
+    "slideshow": {
+     "slide_type": ""
+    },
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "#Define User Interface Function using Gradio and IPython Libraries\n",
+    "import gradio as gr\n",
+    "from IPython.display import Audio\n",
+    "interface = gr.Interface(\n",
+    "    fn=translate_transformers,\n",
+    "    inputs=[\n",
+    "        gr.Dropdown([\"Davlan/byt5-base-eng-yor-mt\", #Exp1\n",
+    "                     \"Davlan/m2m100_418M-eng-yor-mt\", #Exp2\n",
+    "                     \"Davlan/mbart50-large-eng-yor-mt\", #Exp3\n",
+    "                     \"Davlan/mt5_base_eng_yor_mt\", #Exp4\n",
+    "                     \"omoekan/opus-tatoeba-eng-yor\", #Exp5\n",
+    "                     \"masakhane/afrimt5_en_yor_news\", #Exp6\n",
+    "                     \"masakhane/afrimbart_en_yor_news\", #Exp7\n",
+    "                     \"masakhane/afribyt5_en_yor_news\", #Exp8\n",
+    "                     \"masakhane/byt5_en_yor_news\", #Exp9\n",
+    "                     \"masakhane/mt5_en_yor_news\", #Exp10\n",
+    "                     \"masakhane/mbart50_en_yor_news\", #Exp11\n",
+    "                     \"masakhane/m2m100_418M_en_yor_news\", #Exp12\n",
+    "                     \"masakhane/m2m100_418M_en_yor_rel_news\", #Exp13\n",
+    "                     \"masakhane/m2m100_418M_en_yor_rel_news_ft\", #Exp14\n",
+    "                     \"masakhane/m2m100_418M_en_yor_rel\", #Exp15\n",
+    "                     \"dabagyan/menyo_en2yo\", #Exp16\n",
+    "                     #\"facebook/nllb-200-distilled-600M\", #Exp17\n",
+    "                     #\"facebook/nllb-200-3.3B\", #Exp18\n",
+    "                     #\"facebook/nllb-200-1.3B\", #Exp19\n",
+    "                     #\"facebook/nllb-200-distilled-1.3B\",  #Exp20\n",
+    "                     #\"keithhon/nllb-200-3.3B\" #Exp21\n",
+    "                     #\"CohereForAI/aya-101\" #Exp22\n",
+    "                     \"facebook/m2m100_418M\", #Exp17\n",
+    "                     #\"facebook/m2m100_1.2B\",#Exp18\n",
+    "                     #\"facebook/m2m100-12B-avg-5-ckpt\", #Exp19\n",
+    "                     \"google/mt5-base\", #Exp20\n",
+    "                     \"google/byt5-large\" #Exp21\n",
+    "                     ], \n",
+    "                     label=\"Select Finetuned Eng2Yor Translation Model\"),\n",
+    "        gr.Textbox(lines=2, placeholder=\"Enter English Text Here...\", label=\"English Text\")  \n",
+    "    ],\n",
+    "    #outputs = \"text\",\n",
+    "    #outputs=outputs=[\"text\", \"text\"],#\"text\"\n",
+    "    #outputs= gr.Textbox(value=\"text\", label=\"Translated Text\"),\n",
+    "    outputs=[\n",
+    "        gr.Textbox(value=\"text\", label=\"Translated Yoruba Text\"),\n",
+    "        #gr.Textbox(value=\"text\", label=translated_text_actual),\n",
+    "        gr.Textbox(value=\"number\", label=\"BLEU SCORE\"),\n",
+    "        gr.Textbox(value=\"number\", label=\"WER(WORD ERROR RATE) SCORE - The Lower the Better\"),\n",
+    "        gr.Textbox(value=\"number\", label=\"METEOR SCORE\"),\n",
+    "        gr.Textbox(value=\"number\", label=\"MCD(MEL CESPRAL DISTANCE) SCORE\"),\n",
+    "        gr.Audio(type=\"filepath\", label=\"Click to Generate Yoruba Speech from the Translated Text\")\n",
+    "    ],\n",
+    "    title=\"ASPMIR-MACHINE-TRANSLATION-TESTBED FOR LOW RESOURCED AFRICAN LANGUAGES\",\n",
+    "    #gr.Markdown(\"**This Tool Allows Developers and Researchers to Carry Out Experiments on Low Resourced African Languages with State-of-the-Art NMT Finetuned Models.**\"),\n",
+    "    description=\"{This Tool Allows Developers and Researchers to Carry Out Experiments on Low Resourced African Languages with State-of-the-Art Pretrained or Finetuned Models.}\"\n",
+    ")\n",
+    "#interface.launch(share=True)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "c3baee0f-fd85-4209-9d54-14451abd372a",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "* Running on local URL:  http://127.0.0.1:7860\n",
+      "* Running on public URL: https://c18533aae56f5e43a5.gradio.live\n",
+      "\n",
+      "This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from the terminal in the working directory to deploy to Hugging Face Spaces (https://huggingface.co/spaces)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div><iframe src=\"https://c18533aae56f5e43a5.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"
+    }
+   ],
+   "source": [
+    "if __name__ == \"__main__\":\n",
+    "    interface.launch(share=True)"
+   ]
+  }
+ ],
+ "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": "3.10.10"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

ASPMIR-YorTTS.py

@@ -0,0 +1,280 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# In[1]:
+
+
+#Install All the Required Dependencies
+#!pip3 install torch torchvision torchaudio
+#!pip install transformers ipywidgets gradio --upgrade
+#!pip install --upgrade transformers accelerate
+#!pip install --upgrade gradio
+#!pip install nltk
+#!pip install jiwer
+#!pip install sentencepiece
+#!pip install sacremoses
+#!pip install soundfile
+#!pip install librosa numpy jiwer nltk
+#!pip install --upgrade pip 
+#!pip install huggingface_hub
+
+
+# In[2]:
+
+
+#Import Required Libraries
+from transformers import pipeline
+from jiwer import wer
+from transformers import VitsModel, AutoTokenizer, set_seed
+import torch
+import soundfile as sf
+import librosa
+from scipy.spatial.distance import euclidean
+import numpy as np
+import string
+import os
+from nltk.translate.bleu_score import sentence_bleu, SmoothingFunction
+from nltk.translate.meteor_score import meteor_score
+import string
+import numpy as np
+import librosa
+from scipy.spatial.distance import euclidean
+import string
+
+
+# In[3]:
+
+
+import nltk
+nltk.download('wordnet')
+nltk.download('omw-1.4')  # Optional if using WordNet's multilingual features
+import nltk
+print(nltk.data.path)
+import nltk
+nltk.data.path.append('./nltk_data')
+
+
+# In[4]:
+
+
+#Define all Utility Functions
+# Function to compute BLEU score
+def compute_bleu(reference_text, predicted_text):
+    """
+    Computes the BLEU score for a single translation.
+    :param reference_text: The ground truth text (in Yoruba).
+    :param predicted_text: The machine-generated translation text (in Yoruba).
+    :return: BLEU score (float).
+    """
+    print("The Reference Text = ", reference_text)
+    print("The Predicted Text = ",predicted_text)
+    # Tokenize the reference and predicted texts
+    reference_tokens = [reference_text.split()]  # Reference should be wrapped in a list
+    predicted_tokens = predicted_text.split()
+
+    # Add smoothing to handle cases with few n-gram matches
+    smoothing_function = SmoothingFunction().method1
+
+    # Compute BLEU score
+    bleu_score = sentence_bleu(reference_tokens, predicted_tokens, smoothing_function=smoothing_function)
+    #print("The Computed bleu_score in the Compute_Blue Fn = ",bleu_score)
+    return round(bleu_score,2)
+# Function to compute Word Error Rate (WER)
+def compute_wer(reference_text, predicted_text):
+    """
+    Computes the Word Error Rate (WER) for a single translation.
+    :param reference_text: The ground truth text (in Yoruba).
+    :param predicted_text: The machine-generated translation text (in Yoruba).
+    :return: WER score (float).
+    """
+    # Normalize text: lowercase and remove punctuation
+    reference_text = reference_text.lower().translate(str.maketrans('', '', string.punctuation))
+    predicted_text = predicted_text.lower().translate(str.maketrans('', '', string.punctuation))
+
+    # Compute WER
+    wer_score = wer(reference_text, predicted_text)
+
+    return round(wer_score,2)
+
+# Function to compute METEOR score
+def compute_meteor(reference_text, predicted_text):
+    """
+    Computes the METEOR score for a single translation.
+    :param reference_text: The ground truth text (in Yoruba).
+    :param predicted_text: The machine-generated translation text (in Yoruba).
+    :return: METEOR score (float).
+    """
+    # Normalize text: lowercase and remove punctuation
+    reference_text = reference_text.lower().translate(str.maketrans('', '', string.punctuation))
+    predicted_text = predicted_text.lower().translate(str.maketrans('', '', string.punctuation))
+
+    # Tokenize text into lists of words
+    reference_tokens = reference_text.split()
+    predicted_tokens = predicted_text.split()
+
+    # Compute METEOR score
+    meteor = meteor_score([reference_tokens], predicted_tokens)
+    
+    return round(meteor,2)
+
+# Function to compute Mel Cepstral Distance (MCD)
+def compute_mcd(ground_truth_audio_path, predicted_audio_path):
+    """
+    Computes the Mel Cepstral Distance (MCD) between two audio files.
+    :param ground_truth_audio_path: Path to the ground truth audio file.
+    :param predicted_audio_path: Path to the predicted audio file.
+    :return: MCD score (float).
+    """
+    # Load audio files
+    y_true, sr_true = librosa.load(ground_truth_audio_path, sr=16000)
+    y_pred, sr_pred = librosa.load(predicted_audio_path, sr=16000)
+
+    # Ensure the sampling rates match
+    assert sr_true == sr_pred, "Sampling rates do not match between audio files."
+
+    # Compute MFCCs
+    mfcc_true = librosa.feature.mfcc(y=y_true, sr=sr_true, n_mfcc=13).T
+    mfcc_pred = librosa.feature.mfcc(y=y_pred, sr=sr_pred, n_mfcc=13).T
+
+    # Align the MFCC frames
+    min_frames = min(len(mfcc_true), len(mfcc_pred))
+    mfcc_true = mfcc_true[:min_frames]
+    mfcc_pred = mfcc_pred[:min_frames]
+
+    # Compute the Euclidean distance for each frame and average
+    mcd = 0.0
+    for i in range(min_frames):
+        mcd += euclidean(mfcc_true[i], mfcc_pred[i])
+    mcd = (10.0 / np.log(10)) * (mcd / min_frames)
+
+    return round(mcd,2)
+
+
+# In[5]:
+
+
+#Define Translation and Synthesis Function
+def translate_transformers(modelName, sourceLangText):
+    #results = translation_pipeline(input_text)
+    translation_pipeline = pipeline('translation_en_to_yo', model = modelName, max_length=500)
+    translated_text = translation_pipeline(sourceLangText) #translator(text)[0]["translation_text"]
+    translated_text_target = translated_text[0]['translation_text']
+    #reference_translations = "awon apositeli, awon woli, awon ajinrere ati awon oluso agutan ati awon oluko." #'recorder_2024-01-13_11-24-41_453538.wav'#"My name is Joy, I love reading"
+   
+    #TTS for the translated_text_target
+    #TTS Exp1
+    ttsModel = VitsModel.from_pretrained("facebook/mms-tts-yor")
+    tokenizer = AutoTokenizer.from_pretrained("facebook/mms-tts-yor")
+    ttsInputs = tokenizer(translated_text_target, return_tensors="pt")
+    set_seed(555)  # make deterministic
+    with torch.no_grad():
+        ttsOutput = ttsModel(**ttsInputs).waveform
+    #Convert the tensor to a numpy array
+    ttsWaveform = ttsOutput.numpy()[0]    
+    #Save the waveform to an audio file
+    #sf.write('output.wav', waveform, 22050)
+    sf.write('ttsOutput.wav', ttsWaveform, 16000)
+    
+    # Sample ground truth and predicted text2text translations for Clinical Text
+    #ground_truth_text = "Àrùn jẹjẹrẹ ọmú jẹ́ ọ̀kan pàtàkì lára ohun tó ń ṣàkóbá fún ìlera gbogbo ènìyàn ní Nàìjíríà, ó sì jẹ́ ọ̀kan pàtàkì lára ohun tó ń fa ikú àwọn obìnrin tí àrùn jẹjẹrẹ ń pa lórílẹ̀-èdè náà."
+    #predicted_text = translated_text_target #"<extra_id_0> breast cancer is a"
+
+    # Sample ground truth and predicted text2text translations for News Text
+    #ground_truth_text = "Wọ́n ní ìgbà àkọ́kọ́ nìyí tí irú ìwà ipá bẹ́ẹ̀ máa wáyé ní ìpínlẹ̀ Ondo."
+    #predicted_text = translated_text_target #"<extra_id_0> breast cancer is a"
+
+    # Sample ground truth and predicted text2text translations for Religion Text
+    ground_truth_text = "Àwọn aposteli, àwọn wòlíì, àwọn ajíhìnrere, àwọn olùṣọ́-àgùntàn àti àwọn olùkọ́."
+    predicted_text = translated_text_target #"<extra_id_0> breast cancer is a"
+    
+    #Compute bleu_score
+    bleu_score = compute_bleu(ground_truth_text, predicted_text)
+    print(f"Bleu Score (BLEU): {bleu_score:.2f}")
+    
+    #Compute WER
+    wer_score = compute_wer(ground_truth_text, predicted_text)
+    print(f"Word Error Rate (WER): {wer_score:.2f}")
+
+    #Compute METEOR
+    meteor = compute_meteor(ground_truth_text, predicted_text)
+    print(f"METEOR Score: {meteor:.2f}")
+
+    # Paths to sample audio files for MCD computation in current directory
+    ground_truth_audio = os.path.join(os.getcwd(), "gt_ttsOutput.wav")
+    predicted_audio = os.path.join(os.getcwd(), "ttsOutput.wav")
+
+    # Compute Mel Cepstral Distance (MCD)
+    try:
+        mcd = compute_mcd(ground_truth_audio, predicted_audio)
+        print(f"Mel Cepstral Distance (MCD): {mcd:.2f}")
+    except Exception as e:
+        print(f"Error computing MCD: {e}")
+    
+    return translated_text_target,bleu_score,wer_score,meteor,mcd,'ttsOutput.wav'
+
+
+# In[6]:
+
+
+#Define User Interface Function using Gradio and IPython Libraries
+import gradio as gr
+from IPython.display import Audio
+interface = gr.Interface(
+    fn=translate_transformers,
+    inputs=[
+        gr.Dropdown(["Davlan/byt5-base-eng-yor-mt", #Exp1
+                     "Davlan/m2m100_418M-eng-yor-mt", #Exp2
+                     "Davlan/mbart50-large-eng-yor-mt", #Exp3
+                     "Davlan/mt5_base_eng_yor_mt", #Exp4
+                     "omoekan/opus-tatoeba-eng-yor", #Exp5
+                     "masakhane/afrimt5_en_yor_news", #Exp6
+                     "masakhane/afrimbart_en_yor_news", #Exp7
+                     "masakhane/afribyt5_en_yor_news", #Exp8
+                     "masakhane/byt5_en_yor_news", #Exp9
+                     "masakhane/mt5_en_yor_news", #Exp10
+                     "masakhane/mbart50_en_yor_news", #Exp11
+                     "masakhane/m2m100_418M_en_yor_news", #Exp12
+                     "masakhane/m2m100_418M_en_yor_rel_news", #Exp13
+                     "masakhane/m2m100_418M_en_yor_rel_news_ft", #Exp14
+                     "masakhane/m2m100_418M_en_yor_rel", #Exp15
+                     "dabagyan/menyo_en2yo", #Exp16
+                     #"facebook/nllb-200-distilled-600M", #Exp17
+                     #"facebook/nllb-200-3.3B", #Exp18
+                     #"facebook/nllb-200-1.3B", #Exp19
+                     #"facebook/nllb-200-distilled-1.3B",  #Exp20
+                     #"keithhon/nllb-200-3.3B" #Exp21
+                     #"CohereForAI/aya-101" #Exp22
+                     "facebook/m2m100_418M", #Exp17
+                     #"facebook/m2m100_1.2B",#Exp18
+                     #"facebook/m2m100-12B-avg-5-ckpt", #Exp19
+                     "google/mt5-base", #Exp20
+                     "google/byt5-large" #Exp21
+                     ], 
+                     label="Select Finetuned Eng2Yor Translation Model"),
+        gr.Textbox(lines=2, placeholder="Enter English Text Here...", label="English Text")  
+    ],
+    #outputs = "text",
+    #outputs=outputs=["text", "text"],#"text"
+    #outputs= gr.Textbox(value="text", label="Translated Text"),
+    outputs=[
+        gr.Textbox(value="text", label="Translated Yoruba Text"),
+        #gr.Textbox(value="text", label=translated_text_actual),
+        gr.Textbox(value="number", label="BLEU SCORE"),
+        gr.Textbox(value="number", label="WER(WORD ERROR RATE) SCORE - The Lower the Better"),
+        gr.Textbox(value="number", label="METEOR SCORE"),
+        gr.Textbox(value="number", label="MCD(MEL CESPRAL DISTANCE) SCORE"),
+        gr.Audio(type="filepath", label="Click to Generate Yoruba Speech from the Translated Text")
+    ],
+    title="ASPMIR-MACHINE-TRANSLATION-TESTBED FOR LOW RESOURCED AFRICAN LANGUAGES",
+    #gr.Markdown("**This Tool Allows Developers and Researchers to Carry Out Experiments on Low Resourced African Languages with State-of-the-Art NMT Finetuned Models.**"),
+    description="{This Tool Allows Developers and Researchers to Carry Out Experiments on Low Resourced African Languages with State-of-the-Art Pretrained or Finetuned Models.}"
+)
+#interface.launch(share=True)
+
+
+# In[7]:
+
+
+if __name__ == "__main__":
+    interface.launch(share=True)
+

README.md

@@ -1,12 +1,6 @@
 ---
-title: ASPMIR-NMTNEURAL MACHINE TRANSLATION TESTBED
-emoji: 📈
-colorFrom: blue
-colorTo: red
+title: ASPMIR-NMTNEURAL_MACHINE_TRANSLATION_TESTBED
+app_file: ASPMIR-YorTTS.py
 sdk: gradio
-sdk_version: 5.9.1
-app_file: app.py
-pinned: false
+sdk_version: 5.8.0
 ---
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

flagged/log.csv

@@ -0,0 +1,4 @@
+Select Finetuned Eng2Yor Translation Model,English Text,Translated Yoruba Text,WER(Word Error Rate) Score - The Lower the Better,Click to Generate Yoruba Text2Speech,flag,username,timestamp
+masakhane/m2m100_418M_en_yor_news,Recent research has been carried out and revealed that using contraceptives to prevent pregnancy is one of the major causes of breast cancer in the country. All contraceptives sometimes lead to cancer and they are not safe for human consumption,Ìwádìí tí ó ṣẹlẹ̀ ní àìpẹ́ yìí ti fi hàn wípé gbígba ẹ̀dọ̀ tí ó fi dènà àìsàn jẹ́ ọ̀kan lára ohun tí ó ń fa àrùn abẹ ní orílẹ̀-èdè náà. Gbogbo ẹ̀dọ̀ tí ó fi dènà ẹ̀dọ̀ jẹ́ nígbà mìíràn kò sì jẹ́ àbò fún ènìyàn.,1.0,"{""path"":""flagged/Click to Generate Yoruba Text2Speech/68242da6366a05c83761/ttsOutput.wav"",""url"":""https://142dc3e73226f2a320.gradio.live/file=/tmp/gradio/37e7a6078ed13ae05feac2da7f3c0bbe40fb9c91/ttsOutput.wav"",""size"":null,""orig_name"":""ttsOutput.wav"",""mime_type"":null,""is_stream"":false}",,,2024-03-21 13:22:16.576832
+masakhane/m2m100_418M_en_yor_news,How are you,Báwo ni ẹ̀,2.3333333333333335,"{""path"":""flagged/Click to Generate Yoruba Text2Speech/a8120a29323143cfc3fa/ttsOutput.wav"",""url"":""https://142dc3e73226f2a320.gradio.live/file=/tmp/gradio/7625e41f2f7dfe12b18f4ebcb87722ebff349d6d/ttsOutput.wav"",""size"":null,""orig_name"":""ttsOutput.wav"",""mime_type"":null,""is_stream"":false}",,,2024-03-21 13:50:31.879816
+Davlan/m2m100_418M-eng-yor-mt,"My name is Joy, I love programming",text,number,,,,2024-05-17 21:37:06.460794

requirements.txt

@@ -0,0 +1,143 @@
+pip>=24.3.1
+absl-py==1.4.0
+accelerate==1.2.0
+aiofiles==23.2.1
+aiohttp==3.9.0
+aiohttp-cors==0.7.0
+aiosignal==1.3.1
+alembic==1.12.1
+altair==5.2.0
+annotated-types==0.6.0
+anyio==4.0.0
+argon2-cffi==23.1.0
+argon2-cffi-bindings==21.2.0
+array-record==0.5.0
+arrow==1.3.0
+asttokens==2.4.1
+astunparse==1.6.3
+async-generator==1.10
+async-lru==2.0.4
+async-timeout==4.0.3
+attrs==23.1.0
+audioread==3.0.1
+Babel==2.13.1
+beautifulsoup4==4.12.2
+bleach==6.1.0
+blessed==1.20.0
+brotlipy==0.7.0
+cachetools==5.3.2
+certifi
+certipy==0.1.3
+cffi 
+charset-normalizer 
+click==8.1.7
+colorama 
+colorful==0.5.5
+comm==0.2.0
+conda-package-handling 
+conda_package_streaming 
+contourpy==1.2.0
+cryptography 
+cycler==0.12.1
+debugpy==1.8.0
+decorator==5.1.1
+defusedxml==0.7.1
+distlib==0.3.7
+dm-tree==0.1.8
+et-xmlfile==1.1.0
+etils==1.6.0
+exceptiongroup==1.1.3
+executing==2.0.1
+fastapi==0.115.6
+fastjsonschema==2.19.0
+ffmpy==0.3.1
+filelock 
+flatbuffers==23.5.26
+fonttools==4.44.3
+fqdn==1.5.1
+frozenlist==1.4.0
+fsspec==2024.2.0
+gast==0.5.4
+gmpy2 
+google-api-core==2.14.0
+google-auth==2.23.4
+google-auth-oauthlib==1.1.0
+google-pasta==0.2.0
+googleapis-common-protos==1.61.0
+gpustat==1.1.1
+gradio==5.8.0
+gradio_client==1.5.1
+greenlet==3.0.1
+grpcio==1.59.3
+gTTS==2.5.1
+h11==0.14.0
+h5py==3.10.0
+httpcore==1.0.2
+httpx==0.26.0
+huggingface-hub==0.26.3
+idna 
+ipykernel==6.26.0
+ipython==8.17.2
+ipywidgets==8.1.2
+isoduration==20.11.0
+jedi==0.19.1
+Jinja2 
+jiwer==3.0.3
+joblib 
+json5==0.9.14
+jsonpatch 
+jsonpointer 
+jsonschema==4.20.0
+jsonschema-specifications==2023.11.1
+jupyter-events==0.9.0
+jupyter-lsp==2.2.0
+jupyter-resource-usage==1.0.1
+jupyter-telemetry==0.1.0
+jupyter_client==8.6.0
+jupyter_core==5.5.0
+jupyter_server==2.10.1
+jupyter_server_terminals==0.4.4
+jupyterhub==4.0.2
+jupyterlab==4.0.9
+jupyterlab-pygments==0.2.2
+jupyterlab_server==2.25.2
+jupyterlab_widgets==3.0.10
+keras==2.15.0
+librosa==0.10.2.post1
+nltk==3.8.1
+notebook==7.0.6
+notebook_shim==0.2.3
+numba==0.60.0
+numpy 
+nvidia-cublas-cu12==12.1.3.1
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-cuda-nvcc-cu12==12.2.140
+nvidia-cuda-nvrtc-cu12==12.1.105
+nvidia-cuda-runtime-cu12==12.1.105
+nvidia-cudnn-cu12==8.9.2.26
+nvidia-cufft-cu12==11.0.2.54
+nvidia-curand-cu12==10.3.2.106
+nvidia-cusolver-cu12==11.4.5.107
+nvidia-cusparse-cu12==12.1.0.106
+nvidia-ml-py==12.535.133
+nvidia-nccl-cu12==2.20.5
+nvidia-nvjitlink-cu12==12.2.140
+nvidia-nvtx-cu12==12.1.105
+oauthlib==3.2.2
+scapy==2.5.0
+scikit-learn
+scipy 
+SentencePiece
+soundfile==0.12.1
+tensorboard==2.15.1
+tensorboard-data-server==0.7.2
+tensorflow==2.15.0.post1
+tensorflow-estimator==2.15.0
+tensorflow-io-gcs-filesystem==0.34.0
+tensorflow-metadata==1.14.0
+transformers
+torch==2.3.1
+uri-template==1.3.0
+urllib3==2.2.1
+widgetsnbextension==4.0.10
+