Spaces:
Sleeping
Sleeping
File size: 1,629 Bytes
8389164 84ee29a 8389164 f263ba4 8389164 f263ba4 8389164 84ee29a 8389164 84ee29a 8389164 84ee29a d1adb8f 8389164 f263ba4 84ee29a f263ba4 104c95a f263ba4 294ff0d f263ba4 84ee29a f263ba4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 |
from transformers import pipeline
import gradio as gr
import os
model_id = "HarshitJoshi/whisper-small-Hindi"
pipe = pipeline("automatic-speech-recognition", model=model_id)
def transcribe_speech(audio):
output = pipe(
audio,
max_new_tokens=256,
generate_kwargs={
"task": "transcribe",
"language": "hindi",
},
chunk_length_s=10,
batch_size=4,
)
return output["text"]
example_folder = "./examples"
example_files = [f for f in os.listdir(example_folder) if f.endswith('.wav') or f.endswith('.mp3')]
def play_and_transcribe(filename):
filepath = os.path.join(example_folder, filename)
transcription = transcribe_speech(filepath)
return filepath, transcription
with gr.Blocks() as demo:
gr.Markdown("# Hindi Speech Transcription")
with gr.Tab("Transcribe"):
audio_input = gr.Audio(type="filepath", label="Audio Input")
transcribe_button = gr.Button("Transcribe")
output_text = gr.Textbox(label="Transcription")
transcribe_button.click(
fn=transcribe_speech,
inputs=audio_input,
outputs=output_text
)
with gr.Tab("Examples"):
example_dropdown = gr.Dropdown(choices=example_files, label="Select an example")
example_audio = gr.Audio(label="Audio Playback")
example_transcription = gr.Textbox(label="Transcription")
example_dropdown.change(
fn=play_and_transcribe,
inputs=example_dropdown,
outputs=[example_audio, example_transcription]
)
demo.launch(debug=True) |