Spaces:
Sleeping
Sleeping
File size: 2,746 Bytes
d5685b0 5f52293 d5685b0 69f2e98 b1f3cf3 d5685b0 |
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 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 |
import sounddevice as sd
import scipy.io.wavfile as wavfile
import numpy as np
import gradio as gr
from groq import Groq
import tempfile
import os
class Recorder:
def __init__(self, sample_rate=44100):
self.recording = False
self.frames = []
self.sample_rate = sample_rate
self.stream = None
def toggle_recording(self):
if not self.recording:
self.frames = []
self.stream = sd.InputStream(callback=self.callback, channels=2, samplerate=self.sample_rate)
self.stream.start()
self.recording = True
return "Recording... Press to Stop"
else:
self.stream.stop()
self.stream.close()
self.recording = False
return "Recording stopped. Press to Record"
def callback(self, indata, frames, time, status):
if self.recording:
self.frames.append(indata.copy())
def save_audio(self):
if self.frames:
audio_data = np.concatenate(self.frames, axis=0)
with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as temp_wav_file:
wavfile.write(temp_wav_file.name, self.sample_rate, audio_data)
return temp_wav_file.name
else:
return None
recorder = Recorder()
def record():
return recorder.toggle_recording()
def transcribe():
audio_file = recorder.save_audio()
if audio_file:
client = Groq(api_key="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")
with open(audio_file, "rb") as file:
transcription = client.audio.transcriptions.create(
file=(audio_file, file.read()),
model="whisper-large-v3",
prompt="Specify context or spelling", # Optional
response_format="json", # Optional
language="en", # Optional
temperature=0.0 # Optional
)
os.remove(audio_file) # Clean up the temporary file
# Inspect the transcription object to find the text
print(transcription)
# Access the text attribute directly if available
if hasattr(transcription, 'text'):
return transcription.text
else:
return "Transcription text not found."
else:
return "No audio recorded."
with gr.Blocks() as gradio_interface:
with gr.Column():
record_button = gr.Button("Press to Record")
record_button.click(fn=record, outputs=record_button)
transcription_output = gr.Textbox(label="Transcription")
record_button.click(fn=transcribe, outputs=transcription_output)
if __name__ == "__main__":
gradio_interface.launch() |