admin-valentin/WhisperLiveKit
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

bugfix

Browse source
This commit is contained in:
Dominik Machacek 2024-05-28 15:00:11 +02:00
parent 726fa574a2
commit 884958127f
1 changed files with 9 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
whisper_online_server.py
View file

@ -130,21 +130,28 @@ class ServerProcessor:
self.last_end = None self.last_end = None
self.is_first = True
def receive_audio_chunk(self): def receive_audio_chunk(self):
# receive all audio that is available by this time # receive all audio that is available by this time
# blocks operation if less than self.min_chunk seconds is available # blocks operation if less than self.min_chunk seconds is available
# unblocks if connection is closed or a chunk is available # unblocks if connection is closed or a chunk is available
out = [] out = []
while sum(len(x) for x in out) < self.min_chunk*SAMPLING_RATE: minlimit = self.min_chunk*SAMPLING_RATE
while sum(len(x) for x in out) < minlimit:
raw_bytes = self.connection.non_blocking_receive_audio() raw_bytes = self.connection.non_blocking_receive_audio()
print("received audio:",len(raw_bytes), "bytes", raw_bytes[:10])
if not raw_bytes: if not raw_bytes:
break break
print("received audio:",len(raw_bytes), "bytes", raw_bytes[:10])
sf = soundfile.SoundFile(io.BytesIO(raw_bytes), channels=1,endian="LITTLE",samplerate=SAMPLING_RATE, subtype="PCM_16",format="RAW") sf = soundfile.SoundFile(io.BytesIO(raw_bytes), channels=1,endian="LITTLE",samplerate=SAMPLING_RATE, subtype="PCM_16",format="RAW")
audio, _ = librosa.load(sf,sr=SAMPLING_RATE) audio, _ = librosa.load(sf,sr=SAMPLING_RATE)
out.append(audio) out.append(audio)
if not out: if not out:
return None return None
conc = np.concatenate(out)
if self.is_first and len(conc) < minlimit:
return None
self.is_first = False
return np.concatenate(out) return np.concatenate(out)
def format_output_transcript(self,o): def format_output_transcript(self,o):