Upload 4 files

.gitattributes

@@ -0,0 +1,34 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,12 @@
+---
+title: Whisper
+emoji: 🐠
+colorFrom: purple
+colorTo: blue
+sdk: gradio
+sdk_version: 3.12.0
+app_file: app.py
+pinned: false
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,102 @@
+import os
+import gradio as gr
+from transformers import pipeline
+from pytube import YouTube
+from datasets import Dataset, Audio
+from moviepy.editor import AudioFileClip
+import googletrans
+from googletrans import Translator
+
+pipe = pipeline(model="rafat0421/whisper-small-hi")
+
+def download_from_youtube(url):
+    streams = YouTube(url).streams.filter(only_audio=True, file_extension='mp4') #Downloads the video from the given YouTube URL and returns the path to the audio file.
+    fpath = streams.first().download()
+    return fpath
+
+def get_timestamp(seconds):
+    minutes = int(seconds / 60)
+    seconds = int(seconds % 60)
+    return f"{str(minutes).zfill(2)}:{str(seconds).zfill(2)}" #Creates %M:%S timestamp from seconds.
+
+def create_segments(audio_fpath, seconds_max):
+    if not os.path.exists("segmented_audios"):
+        os.makedirs("segmented_audios")
+
+    sound = AudioFileClip(audio_fpath)
+    n_full_segments = int(sound.duration / 30)
+    len_last_segment = sound.duration % 30
+
+    max_segments = int(seconds_max / 30)
+    if n_full_segments > max_segments:
+        n_full_segments = max_segments
+        len_last_segment = 0
+
+    segment_paths = []
+    segment_start_times = []
+
+    segments_available = n_full_segments + 1
+    for i in range(min(segments_available, max_segments)):
+        start = i * 30
+
+        # Skip last segment if it is smaller than two seconds
+        is_last_segment = i == n_full_segments
+        if is_last_segment and not len_last_segment > 2:
+            continue
+        elif is_last_segment:
+            end = start + len_last_segment
+        else:
+            end = (i + 1) * 30
+
+        segment_path = os.path.join("segmented_audios", f"segment_{i}.wav")
+        segment = sound.subclip(start, end)
+        segment.write_audiofile(segment_path)
+        segment_paths.append(segment_path)
+        segment_start_times.append(start)
+
+    return segment_paths, segment_start_times
+
+def get_translation(text):
+    translator = Translator()
+    result = translator.translate(text, src='sv', dest='en')
+
+    return result.text
+
+    #translation_pipeline = pipeline("translation", model=model, tokenizer=tokenizer, src_lang=src_lang, tgt_lang=tgt_lang, max_length=400, device=device)
+    #result = translation_pipeline(text)
+    #return result[0]['translation_text']
+
+    #return "Under Development..."
+
+def transcribe(audio, url, seconds_max):
+    if url:
+        fpath = download_from_youtube(url)
+        segment_paths, segment_start_times = create_segments(fpath, seconds_max)
+
+        audio_dataset = Dataset.from_dict({"audio": segment_paths}).cast_column("audio", Audio(sampling_rate=16000))
+        pred = pipe(audio_dataset["audio"])
+        text = ""
+        n_segments = len(segment_start_times)
+        for i, (seconds, output) in enumerate(zip(segment_start_times, pred)):
+            text += f"[Segment {i+1}/{n_segments}, start time {get_timestamp(seconds)}]\n"
+            text += f"{output['text']}\n"
+            text += f"[Translation]\n{get_translation(output['text'])}\n\n"
+        return text
+
+    else:
+        text = pipe(audio)["text"]
+    return text
+
+iface = gr.Interface(
+    fn=transcribe, 
+    inputs=[
+        gr.Audio(source="microphone", type="filepath", label="Transcribe from Microphone"),
+        gr.Text(max_lines=1, placeholder="YouTube Link", label="Transcribe from YouTube URL"),
+        gr.Slider(minimum=30, maximum=600, value=30, step=30, label="Number of seconds to transcribe")
+    ], 
+    outputs="text",
+    title="Whisper: transcribe Swedish language audio to text",
+    description="Swedish Text Transcription using Transformers.",
+)
+
+iface.launch()

requirements.txt

@@ -0,0 +1,10 @@
+gradio==3.12.0
+transformers==4.25.1
+librosa
+jiwer
+datasets
+torch
+torchaudio
+moviepy
+git+https://github.com/pytube/pytube
+googletrans