dearwolf/aider
1
0
Fork 0
mirror of https://github.com/Aider-AI/aider.git synced 2025-05-29 08:44:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 3

Refactor record_and_transcribe function to remove unnecessary api_key parameter and import dump module.

Browse source
This commit is contained in:
Paul Gauthier 2023-08-10 16:13:23 -03:00
parent 9b526d51e4
commit d74a3ce123
1 changed files with 11 additions and 15 deletions
Download patch file Download diff file Expand all files Collapse all files

26
aider/voice.py
View file

@ -8,11 +8,11 @@ import queue
import soundfile as sf import soundfile as sf
import os import os
def record_and_transcribe(api_key): from .dump import dump
# Set the sample rate and duration for the recording def record_and_transcribe():
sample_rate = 16000 # 16kHz
duration = 10 # in seconds q = queue.Queue()
def callback(indata, frames, time, status): def callback(indata, frames, time, status):
"""This is called (from a separate thread) for each audio block.""" """This is called (from a separate thread) for each audio block."""
@ -23,27 +23,23 @@ def record_and_transcribe(api_key):
filename = tempfile.mktemp(prefix='delme_rec_unlimited_', suffix='.wav', dir='') filename = tempfile.mktemp(prefix='delme_rec_unlimited_', suffix='.wav', dir='')
q = queue.Queue() sample_rate = 16000 # 16kHz
# Make sure the file is opened before recording anything:
with sf.SoundFile(filename, mode='x', samplerate=sample_rate, channels=1) as file: with sf.SoundFile(filename, mode='x', samplerate=sample_rate, channels=1) as file:
with sd.InputStream(samplerate=sample_rate, channels=1, callback=callback): with sd.InputStream(samplerate=sample_rate, channels=1, callback=callback):
input('Press enter when done') input('Press ENTER when done speaking...')
while not q.empty(): while not q.empty():
print('.')
file.write(q.get()) file.write(q.get())
print('done') with open(filename, 'rb') as fh:
transcript = openai.Audio.transcribe("whisper-1", fh)
# Transcribe the audio using the Whisper API text = transcript['text']
response = openai.Whisper.asr.create(audio_data=recording_bytes) return text
# Return the transcription
return response['choices'][0]['text']
if __name__ == "__main__": if __name__ == "__main__":
api_key = os.getenv('OPENAI_API_KEY') api_key = os.getenv('OPENAI_API_KEY')
if not api_key: if not api_key:
raise ValueError("Please set the OPENAI_API_KEY environment variable.") raise ValueError("Please set the OPENAI_API_KEY environment variable.")
print(record_and_transcribe(api_key)) print(record_and_transcribe())