Create app.py

app.py

@@ -0,0 +1,163 @@
+from transformers import pipeline, AutoModelForSpeechSeq2Seq, AutoProcessor
+from transformers.pipelines.audio_utils import ffmpeg_read
+from huggingface_hub import login
+import yt_dlp as youtube_dl
+import gradio as gr
+import tempfile
+import spaces
+import torch
+import time
+import os
+
+login(os.environ["HF"], add_to_git_credential=True)
+
+BATCH_SIZE = 16
+FILE_LIMIT_MB = 1000
+YT_LENGTH_LIMIT_S = 3600  # limit to 1 hour YouTube files
+
+device = "cuda:0" if torch.cuda.is_available() else "cpu"
+torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+model_id = "Kushtrim/whisper-base-shqip"
+model = AutoModelForSpeechSeq2Seq.from_pretrained(
+    model_id, torch_dtype=torch_dtype, use_safetensors=True, token=True).to(device)
+processor = AutoProcessor.from_pretrained(model_id, token=True)
+pipe = pipeline("automatic-speech-recognition", model=model, tokenizer=processor.tokenizer, feature_extractor=processor.feature_extractor,
+                chunk_length_s=30, torch_dtype=torch_dtype, device=device,
+                token=os.environ["HF"])
+
+# pipe = pipeline("automatic-speech-recognition", model=model, tokenizer=processor.tokenizer, feature_extractor=processor.feature_extractor,
+#                 max_new_tokens=128, chunk_length_s=15, batch_size=16, torch_dtype=torch_dtype, device=device,
+#                 token=os.environ["HF"])
+
+@spaces.GPU
+def transcribe(inputs, task):
+    if inputs is None:
+        raise gr.Error(
+            "No audio file submitted! Please upload or record an audio file before submitting your request.")
+
+    text = pipe(inputs, generate_kwargs={
+                "task": task, 'language': 'sq'}, 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()
+
+file_transcribe = gr.Interface(
+    fn=transcribe,
+    inputs=[
+        gr.Audio(sources=["upload"], type="filepath", label="Audio file"),
+        gr.Radio(choices=["transcribe"], label="Task"),
+    ],
+    outputs="text",
+    title="Whisper Large V3 Turbo Shqip: Transcribe Audio",
+    description=(
+        "Easily transcribe long-form audio inputs in Albanian with high accuracy! This demo utilizes the fine-tuned "
+        f"Whisper model [{model_id}](https://huggingface.co/{model_id}), specially adapted for the Albanian language, "
+        "powered by 🤗 Transformers. With just a click, transform microphone or audio file inputs of any length into "
+        "text with exceptional transcription quality."
+    ),
+    allow_flagging="never",
+)
+
+mf_transcribe = gr.Interface(
+    fn=transcribe,
+    inputs=[
+        gr.Audio(sources=["microphone"], type="filepath"),
+        gr.Radio(choices=["transcribe"], label="Task"),
+    ],
+    outputs="text",
+    title="Whisper Large V3 Turbo Shqip: Transcribe Audio",
+    description=(
+        "Easily transcribe long-form audio inputs in Albanian with high accuracy! This demo utilizes the fine-tuned "
+        f"Whisper model [{model_id}](https://huggingface.co/{model_id}), specially adapted for the Albanian language, "
+        "powered by 🤗 Transformers. With just a click, transform microphone or audio file inputs of any length into "
+        "text with exceptional transcription quality."
+    ),
+    allow_flagging="never",
+)
+
+yt_transcribe = gr.Interface(
+    fn=yt_transcribe,
+    inputs=[
+        gr.Textbox(
+            lines=1, placeholder="Paste the URL to a YouTube video here", label="YouTube URL"),
+        gr.Radio(choices=["transcribe"], label="Task")
+    ],
+    outputs=["html", "text"],
+    title="Whisper Large V3 Turbo Shqip: Transcribe Audio",
+    description=(
+        "Easily transcribe long-form audio inputs in Albanian with high accuracy! This demo utilizes the fine-tuned "
+        f"Whisper model [{model_id}](https://huggingface.co/{model_id}), specially adapted for the Albanian language, "
+        "powered by 🤗 Transformers. With just a click, transform microphone or audio file inputs of any length into "
+        "text with exceptional transcription quality."
+    ),
+    allow_flagging="never",
+)
+
+with demo:
+    gr.TabbedInterface([mf_transcribe, file_transcribe, yt_transcribe], ["Microphone", "Audio file", "YouTube"])
+
+demo.launch()