ExpiredPopsicle/whisper-live
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Formatting and spelling fixes.

Browse Source
This commit is contained in:
davabase 2023-10-03 11:21:17 -07:00
parent a0b5bdbb0a
commit f213b229b8
1 changed files with 13 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
transcribe_demo.py
View File

@ -26,33 +26,33 @@ def main():
help="How real time the recording is in seconds.", type=float) help="How real time the recording is in seconds.", type=float)
parser.add_argument("--phrase_timeout", default=3, parser.add_argument("--phrase_timeout", default=3,
help="How much empty space between recordings before we " help="How much empty space between recordings before we "
"consider it a new line in the transcription.", type=float) "consider it a new line in the transcription.", type=float)
if 'linux' in platform: if 'linux' in platform:
parser.add_argument("--default_microphone", default='pulse', parser.add_argument("--default_microphone", default='pulse',
help="Default microphone name for SpeechRecognition. " help="Default microphone name for SpeechRecognition. "
"Run this with 'list' to view available Microphones.", type=str) "Run this with 'list' to view available Microphones.", type=str)
args = parser.parse_args() args = parser.parse_args()
# The last time a recording was retreived from the queue. # The last time a recording was retrieved from the queue.
phrase_time = None phrase_time = None
# Current raw audio bytes. # Current raw audio bytes.
last_sample = bytes() last_sample = bytes()
# Thread safe Queue for passing data from the threaded recording callback. # Thread safe Queue for passing data from the threaded recording callback.
data_queue = Queue() data_queue = Queue()
# We use SpeechRecognizer to record our audio because it has a nice feauture where it can detect when speech ends. # We use SpeechRecognizer to record our audio because it has a nice feature where it can detect when speech ends.
recorder = sr.Recognizer() recorder = sr.Recognizer()
recorder.energy_threshold = args.energy_threshold recorder.energy_threshold = args.energy_threshold
# Definitely do this, dynamic energy compensation lowers the energy threshold dramtically to a point where the SpeechRecognizer never stops recording. # Definitely do this, dynamic energy compensation lowers the energy threshold dramatically to a point where the SpeechRecognizer never stops recording.
recorder.dynamic_energy_threshold = False recorder.dynamic_energy_threshold = False
# Important for linux users. # Important for linux users.
# Prevents permanent application hang and crash by using the wrong Microphone # Prevents permanent application hang and crash by using the wrong Microphone
if 'linux' in platform: if 'linux' in platform:
mic_name = args.default_microphone mic_name = args.default_microphone
if not mic_name or mic_name == 'list': if not mic_name or mic_name == 'list':
print("Available microphone devices are: ") print("Available microphone devices are: ")
for index, name in enumerate(sr.Microphone.list_microphone_names()): for index, name in enumerate(sr.Microphone.list_microphone_names()):
print(f"Microphone with name \"{name}\" found") print(f"Microphone with name \"{name}\" found")
return return
else: else:
for index, name in enumerate(sr.Microphone.list_microphone_names()): for index, name in enumerate(sr.Microphone.list_microphone_names()):
@ -61,7 +61,7 @@ def main():
break break
else: else:
source = sr.Microphone(sample_rate=16000) source = sr.Microphone(sample_rate=16000)
# Load / Download model # Load / Download model
model = args.model model = args.model
if args.model != "large" and not args.non_english: if args.model != "large" and not args.non_english:
@ -73,13 +73,13 @@ def main():
temp_file = NamedTemporaryFile().name temp_file = NamedTemporaryFile().name
transcription = [''] transcription = ['']
with source: with source:
recorder.adjust_for_ambient_noise(source) recorder.adjust_for_ambient_noise(source)
def record_callback(_, audio:sr.AudioData) -> None: def record_callback(_, audio:sr.AudioData) -> None:
""" """
Threaded callback function to recieve audio data when recordings finish. Threaded callback function to receive audio data when recordings finish.
audio: An AudioData containing the recorded bytes. audio: An AudioData containing the recorded bytes.
""" """
# Grab the raw bytes and push it into the thread safe queue. # Grab the raw bytes and push it into the thread safe queue.
@ -124,7 +124,7 @@ def main():
result = audio_model.transcribe(temp_file, fp16=torch.cuda.is_available()) result = audio_model.transcribe(temp_file, fp16=torch.cuda.is_available())
text = result['text'].strip() text = result['text'].strip()
# If we detected a pause between recordings, add a new item to our transcripion. # If we detected a pause between recordings, add a new item to our transcription.
# Otherwise edit the existing one. # Otherwise edit the existing one.
if phrase_complete: if phrase_complete:
transcription.append(text) transcription.append(text)
@ -149,4 +149,4 @@ def main():
if __name__ == "__main__": if __name__ == "__main__":
main() main()