Spaces:
Runtime error
Runtime error
File size: 1,504 Bytes
b625996 aa67f93 b625996 c29a0ae b625996 791702e 26087bc 791702e b625996 2fd0989 791702e 2fd0989 43c3694 705186d 6534ac7 |
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 |
import nltk
import librosa
import torch
import gradio as gr
from transformers import Wav2Vec2Tokenizer, Wav2Vec2ForCTC
nltk.download("punkt")
model_name = "facebook/wav2vec2-base-960h"
tokenizer = Wav2Vec2Tokenizer.from_pretrained(model_name)
model = Wav2Vec2ForCTC.from_pretrained(model_name)
def load_data(input_file):
speech, sample_rate = librosa.load(input_file)
if len(speech.shape) > 1:
speech = speech[:, 0] + speech[:, 1]
if sample_rate != 16000:
speech = librosa.resample(speech, sample_rate, 16000)
return speech
def correct_casing(input_sentence):
sentences = nltk.sent_tokenize(input_sentence)
return (' '.join([s.replace(s[0],s[0].capitalize(),1) for s in sentences]))
def asr_transcript(input_file):
speech = load_data(input_file)
input_values = tokenizer(speech, return_tensors="pt").input_values
logits = model(input_values).logits
predicted_ids = torch.argmax(logits, dim=-1)
transcription = tokenizer.decode(predicted_ids[0])
transcription = correct_casing(transcription.lower())
return transcription
gr.Interface(asr_transcript,
inputs = gr.inputs.Audio(source="microphone", type="filepath", optional=True, label="Speaker"),
outputs = gr.outputs.Textbox(label="Output Text"),
title="ASR using Wav2Vec 2.0",
description = "This application displays transcribed text for given audio input",
examples = [["fr-sb-144.wav"], ["test_File.wav"]], theme="grass").launch()
|