Update app.py

app.py

@@ -9,7 +9,7 @@ from transformers.pipelines.audio_utils import ffmpeg_read
 import tempfile
 import os
 
-MODEL_NAME = "openai/whisper-large-v3-turbo"
+MODEL_NAME = "TalTechNLP/whisper-large-v3-et-subs"
 BATCH_SIZE = 8
 FILE_LIMIT_MB = 1000
 YT_LENGTH_LIMIT_S = 3600  # limit to 1 hour YouTube files
@@ -25,11 +25,11 @@ pipe = pipeline(
 
 
 @spaces.GPU
-def transcribe(inputs, task):
+def transcribe(inputs):
     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, batch_size=BATCH_SIZE, generate_kwargs={"task": task}, return_timestamps=True)["text"]
+    text = pipe(inputs, batch_size=BATCH_SIZE, generate_kwargs={"task": "transcribe", "language": "et"}, return_timestamps=True)["text"]
     return  text
 
 
@@ -73,7 +73,7 @@ def download_yt_audio(yt_url, filename):
             raise gr.Error(str(err))
 
 @spaces.GPU
-def yt_transcribe(yt_url, task, max_filesize=75.0):
+def yt_transcribe(yt_url, max_filesize=75.0):
     html_embed_str = _return_yt_html_embed(yt_url)
 
     with tempfile.TemporaryDirectory() as tmpdirname:
@@ -85,7 +85,7 @@ def yt_transcribe(yt_url, task, max_filesize=75.0):
     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"]
+    text = pipe(inputs, batch_size=BATCH_SIZE, generate_kwargs={"task": "transcribe",  "language": "et"}, return_timestamps=True)["text"]
 
     return html_embed_str, text
 
@@ -95,11 +95,10 @@ demo = gr.Blocks(theme=gr.themes.Ocean())
 mf_transcribe = gr.Interface(
     fn=transcribe,
     inputs=[
-        gr.Audio(sources="microphone", type="filepath"),
-        gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
+        gr.Audio(sources="microphone", type="filepath")
     ],
     outputs="text",
-    title="Whisper Large V3 Turbo: Transcribe Audio",
+    title="Generate Estonian subtitles",
     description=(
         "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the"
         f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe audio files"
@@ -111,11 +110,10 @@ mf_transcribe = gr.Interface(
 file_transcribe = gr.Interface(
     fn=transcribe,
     inputs=[
-        gr.Audio(sources="upload", type="filepath", label="Audio file"),
-        gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
+        gr.Audio(sources="upload", type="filepath", label="Audio file")
     ],
     outputs="text",
-    title="Whisper Large V3: Transcribe Audio",
+    title="Generate Estonian subtitles",
     description=(
         "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the"
         f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe audio files"
@@ -127,11 +125,10 @@ file_transcribe = gr.Interface(
 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(["transcribe", "translate"], label="Task", value="transcribe")
+        gr.Textbox(lines=1, placeholder="Paste the URL to a YouTube video here", label="YouTube URL")
     ],
     outputs=["html", "text"],
-    title="Whisper Large V3: Transcribe YouTube",
+    title="Generate Estonian subtitles",
     description=(
         "Transcribe long-form YouTube videos with the click of a button! Demo uses the checkpoint"
         f" [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe video files of"