initial push

.gitignore

@@ -0,0 +1,81 @@
+# Python build
+.eggs/
+gradio.egg-info
+dist/
+*.pyc
+__pycache__/
+*.py[cod]
+*$py.class
+build/
+__tmp/*
+*.pyi
+py.typed
+
+# JS build
+gradio/templates/*
+gradio/node/*
+gradio/_frontend_code/*
+js/gradio-preview/test/*
+
+# Secrets
+.env
+
+# Gradio run artifacts
+*.db
+*.sqlite3
+gradio/launches.json
+flagged/
+gradio_cached_examples/
+tmp.zip
+
+# Tests
+.coverage
+coverage.xml
+test.txt
+**/snapshots/**/*.png
+playwright-report/
+
+# Demos
+demo/tmp.zip
+demo/files/*.avi
+demo/files/*.mp4
+demo/all_demos/demos/*
+demo/all_demos/requirements.txt
+demo/*/config.json
+demo/annotatedimage_component/*.png
+demo/fake_diffusion_with_gif/*.gif
+
+# Etc
+.idea/*
+.DS_Store
+*.bak
+workspace.code-workspace
+*.h5
+
+# dev containers
+.pnpm-store/
+
+# log files
+.pnpm-debug.log
+
+# Local virtualenv for devs
+venv*
+
+# FRP
+gradio/frpc_*
+.vercel
+
+# js
+node_modules
+public/build/
+test-results
+client/js/test.js
+.config/test.py
+
+# storybook
+storybook-static
+build-storybook.log
+js/storybook/theme.css
+
+# playwright
+.config/playwright/.cache

README.md

@@ -1,13 +1,11 @@
 ---
 title: Whisper Any Model
 emoji: 📉
-colorFrom: indigo
-colorTo: green
+colorFrom: pink
+colorTo: yellow
 sdk: gradio
-sdk_version: 4.22.0
 app_file: app.py
 pinned: false
-license: apache-2.0
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,199 @@
+import torch
+import gradio as gr
+import yt_dlp as youtube_dl
+from transformers import pipeline
+from transformers.pipelines.audio_utils import ffmpeg_read
+
+import tempfile
+import os
+import time
+
+# Available model sizes
+MODEL_CHOICES = ["tiny", "base", "small", "medium", "large", "large-v2", "large-v3"]
+
+current_choice = "tiny"
+DEFAULT_MODEL_NAME = f"openai/whisper-{current_choice}"
+BATCH_SIZE = 8
+FILE_LIMIT_MB = 1000
+YT_LENGTH_LIMIT_S = 3600  # limit to 1 hour YouTube files
+
+device = 0 if torch.cuda.is_available() else "cpu"
+
+# Initialize the pipeline with the default model
+pipe = pipeline(
+    task="automatic-speech-recognition",
+    model=DEFAULT_MODEL_NAME,
+    chunk_length_s=30,
+    device=device,
+)
+
+
+def transcribe(model_size, inputs, task):
+    if inputs is None:
+        raise gr.Error(
+            "No audio file submitted! Please upload or record an audio file before submitting your request."
+        )
+
+    global current_choice
+    global pipe
+
+    current_choice = model_size
+
+    MODEL_NAME = f"openai/whisper-{model_size}"
+    if (
+        pipe.model.name_or_path != MODEL_NAME
+    ):  # Reload the pipeline if model has changed
+        pipe = pipeline(
+            task="automatic-speech-recognition",
+            model=MODEL_NAME,
+            chunk_length_s=30,
+            device=device,
+        )
+
+    text = pipe(
+        inputs,
+        batch_size=BATCH_SIZE,
+        generate_kwargs={"task": task},
+        return_timestamps=True,
+    )["text"]
+    return text
+
+
+def _return_yt_html_embed(yt_url):
+    video_id = yt_url.split("?v=")[-1]
+    HTML_str = (
+        f'<center> <iframe width="500" height="320" src="https://www.youtube.com/embed/{video_id}"> </iframe>'
+        " </center>"
+    )
+    return HTML_str
+
+
+def download_yt_audio(yt_url, filename):
+    info_loader = youtube_dl.YoutubeDL()
+
+    try:
+        info = info_loader.extract_info(yt_url, download=False)
+    except youtube_dl.utils.DownloadError as err:
+        raise gr.Error(str(err))
+
+    file_length = info["duration_string"]
+    file_h_m_s = file_length.split(":")
+    file_h_m_s = [int(sub_length) for sub_length in file_h_m_s]
+
+    if len(file_h_m_s) == 1:
+        file_h_m_s.insert(0, 0)
+    if len(file_h_m_s) == 2:
+        file_h_m_s.insert(0, 0)
+    file_length_s = file_h_m_s[0] * 3600 + file_h_m_s[1] * 60 + file_h_m_s[2]
+
+    if file_length_s > YT_LENGTH_LIMIT_S:
+        yt_length_limit_hms = time.strftime(
+            "%HH:%MM:%SS", time.gmtime(YT_LENGTH_LIMIT_S)
+        )
+        file_length_hms = time.strftime("%HH:%MM:%SS", time.gmtime(file_length_s))
+        raise gr.Error(
+            f"Maximum YouTube length is {yt_length_limit_hms}, got {file_length_hms} YouTube video."
+        )
+
+    ydl_opts = {
+        "outtmpl": filename,
+        "format": "worstvideo[ext=mp4]+bestaudio[ext=m4a]/best[ext=mp4]/best",
+    }
+
+    with youtube_dl.YoutubeDL(ydl_opts) as ydl:
+        try:
+            ydl.download([yt_url])
+        except youtube_dl.utils.ExtractorError as err:
+            raise gr.Error(str(err))
+
+
+def yt_transcribe(yt_url, task, max_filesize=75.0):
+    html_embed_str = _return_yt_html_embed(yt_url)
+
+    with tempfile.TemporaryDirectory() as tmpdirname:
+        filepath = os.path.join(tmpdirname, "video.mp4")
+        download_yt_audio(yt_url, filepath)
+        with open(filepath, "rb") as f:
+            inputs = f.read()
+
+    inputs = ffmpeg_read(inputs, pipe.feature_extractor.sampling_rate)
+    inputs = {"array": inputs, "sampling_rate": pipe.feature_extractor.sampling_rate}
+
+    text = pipe(
+        inputs,
+        batch_size=BATCH_SIZE,
+        generate_kwargs={"task": task},
+        return_timestamps=True,
+    )["text"]
+
+    return html_embed_str, text
+
+
+demo = gr.Blocks()
+
+mf_transcribe = gr.Interface(
+    fn=transcribe,
+    inputs=[
+        gr.Dropdown(MODEL_CHOICES, label="Model Size", value=current_choice),
+        gr.Audio(sources=["microphone"], type="filepath"),
+        gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
+    ],
+    outputs="text",
+    theme="default",
+    title="Whisper: Transcribe Audio",
+    description=(
+        "Transcribe long-form microphone or audio inputs with the click of a button! Demo allows selection of any of the"
+        f" [OpenAI Whisper model sizes](https://huggingface.co/openai/whisper-large-v3) and  Transformers to transcribe audio files"
+        " of arbitrary length. Large and above are multilingual."
+        " Based on https://huggingface.co/spaces/openai/whisper"
+    ),
+    allow_flagging="never",
+)
+
+file_transcribe = gr.Interface(
+    fn=transcribe,
+    inputs=[
+        gr.Dropdown(MODEL_CHOICES, label="Model Size", value=current_choice),
+        gr.Audio(sources=["upload"], type="filepath", label="Audio file"),
+        gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
+    ],
+    outputs="text",
+    theme="default",
+    title="Whisper: Transcribe Audio",
+    description=(
+        "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the OpenAI Whisper"
+        f" checkpoint [{DEFAULT_MODEL_NAME}](https://huggingface.co/{DEFAULT_MODEL_NAME}) and  Transformers to transcribe audio files"
+        " of arbitrary length."
+    ),
+    allow_flagging="never",
+)
+
+yt_transcribe = gr.Interface(
+    fn=yt_transcribe,
+    inputs=[
+        gr.Dropdown(MODEL_CHOICES, label="Model Size", value=current_choice),
+        gr.Textbox(
+            lines=1,
+            placeholder="Paste the URL to a YouTube video here",
+            label="YouTube URL",
+        ),
+        gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
+    ],
+    outputs=["html", "text"],
+    theme="default",
+    title="Whisper: Transcribe Audio",
+    description=(
+        "Transcribe long-form YouTube videos with the click of a button! Demo uses the OpenAI Whisper checkpoint"
+        f" [{DEFAULT_MODEL_NAME}](https://huggingface.co/{DEFAULT_MODEL_NAME}) and  Transformers to transcribe video files of"
+        " arbitrary length."
+    ),
+    allow_flagging="never",
+)
+
+with demo:
+    gr.TabbedInterface(
+        [mf_transcribe, file_transcribe, yt_transcribe],
+        ["Microphone", "Audio file", "YouTube"],
+    )
+
+demo.launch()

requirements.txt

@@ -0,0 +1,4 @@
+gradio
+git+https://github.com/huggingface/transformers
+torch
+yt-dlp