Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -1,6 +1,7 @@
|
|
1 |
-
import io
|
2 |
import os
|
|
|
3 |
import gradio as gr
|
|
|
4 |
from google.cloud import speech
|
5 |
from google.api_core.client_options import ClientOptions
|
6 |
|
@@ -11,11 +12,25 @@ API_KEY = os.getenv("GOOGLE_API_KEY")
|
|
11 |
if not API_KEY:
|
12 |
raise ValueError("La API Key de Google no está configurada. Configúrala en la variable de entorno GOOGLE_API_KEY.")
|
13 |
|
14 |
-
def
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
"""Transcribe audio a texto usando Google Cloud Speech-to-Text con API Key."""
|
16 |
-
if
|
17 |
return '', ''
|
18 |
|
|
|
|
|
|
|
|
|
19 |
# Configurar el cliente de Speech-to-Text con API Key
|
20 |
client_options = ClientOptions(api_key=API_KEY)
|
21 |
client = speech.SpeechClient(client_options=client_options)
|
@@ -29,7 +44,7 @@ def transcribe(file_name):
|
|
29 |
)
|
30 |
|
31 |
# Cargar el audio en binario
|
32 |
-
with io.open(
|
33 |
content = audio_file.read()
|
34 |
audio = speech.RecognitionAudio(content=content)
|
35 |
|
@@ -46,16 +61,19 @@ def transcribe(file_name):
|
|
46 |
|
47 |
return ' '.join(transcript), '\n'.join(confidence)
|
48 |
|
49 |
-
# Configuración de la interfaz Gradio
|
50 |
output1 = gr.Textbox(label='Transcripción')
|
51 |
output2 = gr.Textbox(label='Confianza')
|
52 |
|
53 |
demo = gr.Interface(
|
54 |
transcribe,
|
55 |
-
|
|
|
|
|
|
|
56 |
[output1, output2],
|
57 |
-
title='Demo Reconocimiento de
|
58 |
-
description='<p>Grabar audio para convertir voz a texto usando
|
59 |
)
|
60 |
|
61 |
demo.launch()
|
|
|
|
|
1 |
import os
|
2 |
+
import io
|
3 |
import gradio as gr
|
4 |
+
import subprocess
|
5 |
from google.cloud import speech
|
6 |
from google.api_core.client_options import ClientOptions
|
7 |
|
|
|
12 |
if not API_KEY:
|
13 |
raise ValueError("La API Key de Google no está configurada. Configúrala en la variable de entorno GOOGLE_API_KEY.")
|
14 |
|
15 |
+
def convert_to_wav(input_file):
|
16 |
+
"""Convierte archivos de audio a formato WAV LINEAR16 si es necesario."""
|
17 |
+
output_file = input_file + ".wav"
|
18 |
+
command = [
|
19 |
+
"ffmpeg", "-y", "-i", input_file,
|
20 |
+
"-acodec", "pcm_s16le", "-ar", "44100", "-ac", "1", output_file
|
21 |
+
]
|
22 |
+
subprocess.run(command, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
|
23 |
+
return output_file
|
24 |
+
|
25 |
+
def transcribe(file_path):
|
26 |
"""Transcribe audio a texto usando Google Cloud Speech-to-Text con API Key."""
|
27 |
+
if file_path is None:
|
28 |
return '', ''
|
29 |
|
30 |
+
# Convertir a formato WAV si es necesario
|
31 |
+
if not file_path.endswith(".wav"):
|
32 |
+
file_path = convert_to_wav(file_path)
|
33 |
+
|
34 |
# Configurar el cliente de Speech-to-Text con API Key
|
35 |
client_options = ClientOptions(api_key=API_KEY)
|
36 |
client = speech.SpeechClient(client_options=client_options)
|
|
|
44 |
)
|
45 |
|
46 |
# Cargar el audio en binario
|
47 |
+
with io.open(file_path, "rb") as audio_file:
|
48 |
content = audio_file.read()
|
49 |
audio = speech.RecognitionAudio(content=content)
|
50 |
|
|
|
61 |
|
62 |
return ' '.join(transcript), '\n'.join(confidence)
|
63 |
|
64 |
+
# Configuración de la interfaz Gradio con opciones de grabación y subida de archivos
|
65 |
output1 = gr.Textbox(label='Transcripción')
|
66 |
output2 = gr.Textbox(label='Confianza')
|
67 |
|
68 |
demo = gr.Interface(
|
69 |
transcribe,
|
70 |
+
[
|
71 |
+
gr.Audio(sources=["microphone"], type="filepath", label='Grabar audio aquí', streaming=False),
|
72 |
+
gr.File(label="Subir archivo de audio")
|
73 |
+
],
|
74 |
[output1, output2],
|
75 |
+
title='Demo Reconocimiento de Voz con Google',
|
76 |
+
description='<p>Grabar o subir un archivo de audio para convertir voz a texto usando Google Cloud Speech-to-Text.</p>'
|
77 |
)
|
78 |
|
79 |
demo.launch()
|