Hunzla commited on
Commit
f84f4f1
·
1 Parent(s): 164d655

Create app.py

Browse files
Files changed (1) hide show
  1. app.py +33 -0
app.py ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from transformers import pipeline
2
+ from gradio_client import Client
3
+ asr_pipe = pipeline("automatic-speech-recognition", model="Abdullah17/whisper-small-urdu")
4
+
5
+ def transcribe_the_command(audio):
6
+ import soundfile as sf
7
+ sample_rate, audio_data = audio
8
+ file_name = "recorded_audio.wav"
9
+ sf.write(file_name, audio_data, sample_rate)
10
+ # Convert stereo to mono by averaging the two channels
11
+ print(file_name)
12
+
13
+ transcript = asr_pipe(file_name)["text"]
14
+ client = Client("https://ysharma-explore-llamav2-with-tgi.hf.space/")
15
+ result = client.predict(
16
+ "tell me a story",
17
+ api_name="/chat"
18
+ )
19
+
20
+ return result
21
+ # get_text_from_voice("urdu.wav")
22
+ import gradio as gr
23
+
24
+
25
+ iface = gr.Interface(
26
+ fn=transcribe_the_command,
27
+ inputs=gr.inputs.Audio(label="Recorded Audio",source="microphone"),
28
+ outputs="text",
29
+ title="Whisper Small Urdu Command",
30
+ description="Realtime demo for Urdu speech recognition using a fine-tuned Whisper small model and outputting the estimated command on the basis of speech transcript.",
31
+ )
32
+
33
+ iface.launch()