MMS_ASR_Inference__English_Colab .ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "machine_shape": "hm",
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# 1.Clone fairseq-py and install latest version, other packages for downloading youtube video\n"
      ],
      "metadata": {
        "id": "2XlbnCU8360p"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zAAHe9sS6dMg",
        "outputId": "ff70838a-91ca-4822-f269-4c3e04f95a74"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting pytube\n",
            "  Downloading pytube-15.0.0-py3-none-any.whl (57 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m57.6/57.6 kB\u001b[0m \u001b[31m2.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: moviepy in /usr/local/lib/python3.10/dist-packages (1.0.3)\n",
            "Requirement already satisfied: decorator<5.0,>=4.0.2 in /usr/local/lib/python3.10/dist-packages (from moviepy) (4.4.2)\n",
            "Requirement already satisfied: tqdm<5.0,>=4.11.2 in /usr/local/lib/python3.10/dist-packages (from moviepy) (4.65.0)\n",
            "Requirement already satisfied: requests<3.0,>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from moviepy) (2.27.1)\n",
            "Requirement already satisfied: proglog<=1.0.0 in /usr/local/lib/python3.10/dist-packages (from moviepy) (0.1.10)\n",
            "Requirement already satisfied: numpy>=1.17.3 in /usr/local/lib/python3.10/dist-packages (from moviepy) (1.22.4)\n",
            "Requirement already satisfied: imageio<3.0,>=2.5 in /usr/local/lib/python3.10/dist-packages (from moviepy) (2.25.1)\n",
            "Requirement already satisfied: imageio-ffmpeg>=0.2.0 in /usr/local/lib/python3.10/dist-packages (from moviepy) (0.4.8)\n",
            "Requirement already satisfied: pillow>=8.3.2 in /usr/local/lib/python3.10/dist-packages (from imageio<3.0,>=2.5->moviepy) (8.4.0)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests<3.0,>=2.8.1->moviepy) (1.26.15)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests<3.0,>=2.8.1->moviepy) (2022.12.7)\n",
            "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.10/dist-packages (from requests<3.0,>=2.8.1->moviepy) (2.0.12)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests<3.0,>=2.8.1->moviepy) (3.4)\n",
            "Installing collected packages: pytube\n",
            "Successfully installed pytube-15.0.0\n"
          ]
        }
      ],
      "source": [
        " !pip install pytube moviepy"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!mkdir \"temp_dir\"\n",
        "!git clone https://github.com/pytorch/fairseq\n",
        "\n",
        "# Change current working directory\n",
        "!pwd\n",
        "%cd \"/content/fairseq\"\n",
        "!pip install --editable ./ \n",
        "!pip install tensorboardX\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uJSTZKg_9JcN",
        "outputId": "003ce36b-e37c-4ab7-91d7-822823223ad4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cloning into 'fairseq'...\n",
            "remote: Enumerating objects: 34719, done.\u001b[K\n",
            "remote: Counting objects: 100% (176/176), done.\u001b[K\n",
            "remote: Compressing objects: 100% (103/103), done.\u001b[K\n",
            "remote: Total 34719 (delta 87), reused 142 (delta 68), pack-reused 34543\u001b[K\n",
            "Receiving objects: 100% (34719/34719), 25.00 MiB | 26.64 MiB/s, done.\n",
            "Resolving deltas: 100% (25179/25179), done.\n",
            "/content\n",
            "/content/fairseq\n",
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Obtaining file:///content/fairseq\n",
            "  Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n",
            "  Checking if build backend supports build_editable ... \u001b[?25l\u001b[?25hdone\n",
            "  Getting requirements to build editable ... \u001b[?25l\u001b[?25hdone\n",
            "  Preparing editable metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: cffi in /usr/local/lib/python3.10/dist-packages (from fairseq==0.12.2) (1.15.1)\n",
            "Requirement already satisfied: cython in /usr/local/lib/python3.10/dist-packages (from fairseq==0.12.2) (0.29.34)\n",
            "Collecting hydra-core<1.1,>=1.0.7 (from fairseq==0.12.2)\n",
            "  Downloading hydra_core-1.0.7-py3-none-any.whl (123 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m123.8/123.8 kB\u001b[0m \u001b[31m3.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting omegaconf<2.1 (from fairseq==0.12.2)\n",
            "  Downloading omegaconf-2.0.6-py3-none-any.whl (36 kB)\n",
            "Requirement already satisfied: numpy>=1.21.3 in /usr/local/lib/python3.10/dist-packages (from fairseq==0.12.2) (1.22.4)\n",
            "Requirement already satisfied: regex in /usr/local/lib/python3.10/dist-packages (from fairseq==0.12.2) (2022.10.31)\n",
            "Collecting sacrebleu>=1.4.12 (from fairseq==0.12.2)\n",
            "  Downloading sacrebleu-2.3.1-py3-none-any.whl (118 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m118.9/118.9 kB\u001b[0m \u001b[31m13.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: torch>=1.13 in /usr/local/lib/python3.10/dist-packages (from fairseq==0.12.2) (2.0.1+cu118)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from fairseq==0.12.2) (4.65.0)\n",
            "Collecting bitarray (from fairseq==0.12.2)\n",
            "  Downloading bitarray-2.7.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (273 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m273.7/273.7 kB\u001b[0m \u001b[31m15.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: torchaudio>=0.8.0 in /usr/local/lib/python3.10/dist-packages (from fairseq==0.12.2) (2.0.2+cu118)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.10/dist-packages (from fairseq==0.12.2) (1.2.2)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from fairseq==0.12.2) (23.1)\n",
            "Collecting antlr4-python3-runtime==4.8 (from hydra-core<1.1,>=1.0.7->fairseq==0.12.2)\n",
            "  Downloading antlr4-python3-runtime-4.8.tar.gz (112 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m112.4/112.4 kB\u001b[0m \u001b[31m15.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: PyYAML>=5.1.* in /usr/local/lib/python3.10/dist-packages (from omegaconf<2.1->fairseq==0.12.2) (6.0)\n",
            "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.10/dist-packages (from omegaconf<2.1->fairseq==0.12.2) (4.5.0)\n",
            "Collecting portalocker (from sacrebleu>=1.4.12->fairseq==0.12.2)\n",
            "  Downloading portalocker-2.7.0-py2.py3-none-any.whl (15 kB)\n",
            "Requirement already satisfied: tabulate>=0.8.9 in /usr/local/lib/python3.10/dist-packages (from sacrebleu>=1.4.12->fairseq==0.12.2) (0.8.10)\n",
            "Collecting colorama (from sacrebleu>=1.4.12->fairseq==0.12.2)\n",
            "  Downloading colorama-0.4.6-py2.py3-none-any.whl (25 kB)\n",
            "Requirement already satisfied: lxml in /usr/local/lib/python3.10/dist-packages (from sacrebleu>=1.4.12->fairseq==0.12.2) (4.9.2)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch>=1.13->fairseq==0.12.2) (3.12.0)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch>=1.13->fairseq==0.12.2) (1.11.1)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch>=1.13->fairseq==0.12.2) (3.1)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch>=1.13->fairseq==0.12.2) (3.1.2)\n",
            "Requirement already satisfied: triton==2.0.0 in /usr/local/lib/python3.10/dist-packages (from torch>=1.13->fairseq==0.12.2) (2.0.0)\n",
            "Requirement already satisfied: cmake in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch>=1.13->fairseq==0.12.2) (3.25.2)\n",
            "Requirement already satisfied: lit in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch>=1.13->fairseq==0.12.2) (16.0.5)\n",
            "Requirement already satisfied: pycparser in /usr/local/lib/python3.10/dist-packages (from cffi->fairseq==0.12.2) (2.21)\n",
            "Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.10/dist-packages (from scikit-learn->fairseq==0.12.2) (1.10.1)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn->fairseq==0.12.2) (1.2.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn->fairseq==0.12.2) (3.1.0)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch>=1.13->fairseq==0.12.2) (2.1.2)\n",
            "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch>=1.13->fairseq==0.12.2) (1.3.0)\n",
            "Building wheels for collected packages: fairseq, antlr4-python3-runtime\n",
            "  Building editable for fairseq (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for fairseq: filename=fairseq-0.12.2-0.editable-cp310-cp310-linux_x86_64.whl size=9219 sha256=4a90f1ec156a635bef4c243cbe492df71a4071cac8afc976b885560760462a3b\n",
            "  Stored in directory: /tmp/pip-ephem-wheel-cache-_dbu3lm7/wheels/c6/d7/db/bc419b1daa8266aa8de2a7c4d29f62dbfa814e8701fe4695a2\n",
            "  Building wheel for antlr4-python3-runtime (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for antlr4-python3-runtime: filename=antlr4_python3_runtime-4.8-py3-none-any.whl size=141210 sha256=12fd676964c81c309ac329a88abfa7cb60b6ec756f07a49036782c385c6b899f\n",
            "  Stored in directory: /root/.cache/pip/wheels/a7/20/bd/e1477d664f22d99989fd28ee1a43d6633dddb5cb9e801350d5\n",
            "Successfully built fairseq antlr4-python3-runtime\n",
            "Installing collected packages: bitarray, antlr4-python3-runtime, portalocker, omegaconf, colorama, sacrebleu, hydra-core, fairseq\n",
            "Successfully installed antlr4-python3-runtime-4.8 bitarray-2.7.4 colorama-0.4.6 fairseq-0.12.2 hydra-core-1.0.7 omegaconf-2.0.6 portalocker-2.7.0 sacrebleu-2.3.1\n",
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting tensorboardX\n",
            "  Downloading tensorboardX-2.6-py2.py3-none-any.whl (114 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m114.5/114.5 kB\u001b[0m \u001b[31m3.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from tensorboardX) (1.22.4)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from tensorboardX) (23.1)\n",
            "Requirement already satisfied: protobuf<4,>=3.8.0 in /usr/local/lib/python3.10/dist-packages (from tensorboardX) (3.20.3)\n",
            "Installing collected packages: tensorboardX\n",
            "Successfully installed tensorboardX-2.6\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 2.Download MMS model"
      ],
      "metadata": {
        "id": "10QltDsO3t9B"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Un-comment to download your preferred model. In this video, I use MMS-1B-all for better model quality and language coverage, (but it would require more RAM, so please use Colab-Pro instead of Colab-Free). You can use Colab-Free for MMS-1B:FL10 or MMS-1B:L1107"
      ],
      "metadata": {
        "id": "TK_MmMdY4oaY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# MMS-1B:FL102 model - 102 Languages - FLEURS Dataset\n",
        "#!wget -P ./models_new 'https://dl.fbaipublicfiles.com/mms/asr/mms1b_fl102.pt'\n",
        "\n",
        "# # MMS-1B:L1107 - 1107 Languages - MMS-lab Dataset\n",
        "# !wget -P ./models_new 'https://dl.fbaipublicfiles.com/mms/asr/mms1b_l1107.pt'\n",
        "\n",
        "# # MMS-1B-all - 1162 Languages - MMS-lab + FLEURS + CV + VP + MLS\n",
        "!wget -P ./models_new 'https://dl.fbaipublicfiles.com/mms/asr/mms1b_all.pt'"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NHEMEGlr9bcq",
        "outputId": "253768b9-a165-41ed-a4b8-715c7b457109"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2023-06-03 19:21:12--  https://dl.fbaipublicfiles.com/mms/asr/mms1b_all.pt\n",
            "Resolving dl.fbaipublicfiles.com (dl.fbaipublicfiles.com)... 108.157.162.120, 108.157.162.108, 108.157.162.35, ...\n",
            "Connecting to dl.fbaipublicfiles.com (dl.fbaipublicfiles.com)|108.157.162.120|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 14660831159 (14G) [binary/octet-stream]\n",
            "Saving to: ‘./models_new/mms1b_all.pt’\n",
            "\n",
            "mms1b_all.pt        100%[===================>]  13.65G   156MB/s    in 93s     \n",
            "\n",
            "2023-06-03 19:22:45 (150 MB/s) - ‘./models_new/mms1b_all.pt’ saved [14660831159/14660831159]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 3.Download youtube CNN video"
      ],
      "metadata": {
        "id": "QbFxkAmo52jV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from pytube import YouTube\n",
        "from moviepy.editor import *\n",
        "import os\n",
        "\n",
        "# replace 'your_youtube_url_here' with your YouTube video's URL\n",
        "youtube_url = 'https://www.youtube.com/watch?v=3dRQ9JApJbs'  \n",
        "os.makedirs('/content/audio_chunks2', exist_ok=True)\n",
        "\n",
        "# download video\n",
        "youtube = YouTube(youtube_url)\n",
        "video = youtube.streams.first()\n",
        "filename = video.download()\n",
        "\n",
        "# convert video to audio\n",
        "video_clip = VideoFileClip(filename)\n",
        "audio_filename = filename.split('.')[0] + '.wav'\n",
        "video_clip.audio.write_audiofile(audio_filename)\n",
        "\n",
        "# split audio into 1-minute chunks\n",
        "audio_clip = AudioFileClip(audio_filename)\n",
        "duration = int(audio_clip.duration)\n",
        "\n",
        "# Initialize a counter\n",
        "counter = 0\n",
        "\n",
        "for i in range(0, duration, 60):  # we'll split the audio every 60 seconds\n",
        "    start = i\n",
        "    end = i + 60 if i + 60 < duration else duration  # make sure not to go over the video's duration\n",
        "    subclip = audio_clip.subclip(start, end)\n",
        "    # Include the counter in the filename\n",
        "    subclip_filename = f\"/content/audio_chunks2/{os.path.basename(audio_filename).split('.')[0]}_{str(counter).zfill(5)}.wav\"\n",
        "    subclip.write_audiofile(subclip_filename)\n",
        "    \n",
        "    # Increment the counter\n",
        "    counter += 1\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lp8LEndYumup",
        "outputId": "677925c3-1d5d-4d5d-8ba2-d41f542ca5b4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MoviePy - Writing audio in /content/fairseq/Unprecedented images show sunspot larger than Earth.wav\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": []
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MoviePy - Done.\n",
            "MoviePy - Writing audio in /content/audio_chunks2/Unprecedented images show sunspot larger than Earth_00000.wav\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": []
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MoviePy - Done.\n",
            "MoviePy - Writing audio in /content/audio_chunks2/Unprecedented images show sunspot larger than Earth_00001.wav\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": []
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MoviePy - Done.\n",
            "MoviePy - Writing audio in /content/audio_chunks2/Unprecedented images show sunspot larger than Earth_00002.wav\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": []
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MoviePy - Done.\n",
            "MoviePy - Writing audio in /content/audio_chunks2/Unprecedented images show sunspot larger than Earth_00003.wav\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": []
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MoviePy - Done.\n",
            "MoviePy - Writing audio in /content/audio_chunks2/Unprecedented images show sunspot larger than Earth_00004.wav\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "                                                                     "
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MoviePy - Done.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\r"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# to keep the order sme\n",
        "import os\n",
        "import glob\n",
        "import subprocess\n",
        "\n",
        "# Define your directory\n",
        "directory = '/content/audio_chunks2/'\n",
        "\n",
        "# Use glob to get all the .wav files in the directory\n",
        "input_files = glob.glob(os.path.join(directory, '*.wav'))\n",
        "\n",
        "# Sort the files before processing\n",
        "input_files = sorted(input_files)\n",
        "\n",
        "# Loop through the input files\n",
        "for i, input_filename in enumerate(input_files):\n",
        "    # Define the output filename using a shorter name\n",
        "    output_filename = os.path.join(directory, f'res_{i}.wav')\n",
        "\n",
        "    # Run the ffmpeg command to resample and save with the new filename\n",
        "    # convert to 16000 hz sample rate\n",
        "    subprocess.run(['ffmpeg', '-y', '-i', input_filename, '-ar', '16000', output_filename])\n"
      ],
      "metadata": {
        "id": "TJSZAmshN9n4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 4.Run Inference and transcribe the audio(s)"
      ],
      "metadata": {
        "id": "OjCm7qGAUWw3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "import os\n",
        "import subprocess\n",
        "\n",
        "os.environ[\"TMPDIR\"] = '/content/temp_dir'\n",
        "os.environ[\"PYTHONPATH\"] = \".\"\n",
        "os.environ[\"PREFIX\"] = \"INFER\"\n",
        "os.environ[\"HYDRA_FULL_ERROR\"] = \"1\"\n",
        "os.environ[\"USER\"] = \"micro\"\n",
        "\n",
        "# Define your directory and the base file name\n",
        "directory = '/content/audio_chunks2/'\n",
        "base_filename = 'res_'\n",
        "\n",
        "# Define the range of file indices\n",
        "file_indices = range(5)  # This will give you the numbers 0 to 5\n",
        "\n",
        "# Initialize an empty string to hold the full transcription\n",
        "full_transcription = \"\"\n",
        "\n",
        "# Loop through the file indices\n",
        "for i in file_indices:\n",
        "    # Construct the input file name\n",
        "    input_filename = os.path.join(directory, f'{base_filename}{i}.wav')\n",
        "\n",
        "    # Define the output filename for the transcription\n",
        "    transcription_filename = f'transcription_{i}.txt'\n",
        "\n",
        "    # Open the transcription file in write mode\n",
        "    with open(transcription_filename, 'w') as f:\n",
        "        # Run the model inference and save the transcription to the file        # Replace mms1b_all.pt to mms1b_fl102.pt if you want to use a different model\n",
        "                                                                                #https://dl.fbaipublicfiles.com/mms/asr/mms1b_all_langs.html, if you use different language\n",
        "        subprocess.run(['python', 'examples/mms/asr/infer/mms_infer.py', '--model', '/content/fairseq/models_new/mmps1b_all.pt', '--lang', 'eng', '--audio', input_filename], stdout=f)\n",
        "\n",
        "    # Open the transcription file in read mode and append its contents to the full transcription\n",
        "    with open(transcription_filename, 'r') as f:\n",
        "        transcription = f.read()\n",
        "        full_transcription += \" \" + transcription\n",
        "\n",
        "# Save the full transcription to a file in the specified directory\n",
        "with open('/content/audio_chunks2/full_transcription.txt', 'w') as f:\n",
        "    f.write(full_transcription)"
      ],
      "metadata": {
        "id": "1qA642MsBgTv"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}