Trying to detect speech using VAD(Voice Activity Detector)

Question

I am able to read the audio but I am getting an error message while passing it to VAD(Voice Activity Detector). I think the error message is because the frames is in bytes, when feeding it to vad.is_speech(frame, sample_rate), should this frame be in bytes? Here is the code below:

frame_duration_ms=10
duration_in_ms = (frame_duration_ms / 1000) #duration in 10ms
frame_size = int(sample_rate * duration_in_ms) #frame size of 160
frame_bytes = frame_size * 2

def frame_generator(buffer, frame_bytes):
    # repeatedly store 320 length array to the frame_stored when the frame_bytes is less than the size of the buffer
    while offset+frame_bytes < len(buffer):
        frame_stored = buffer[offset : offset+frame_bytes]
        offset = offset + frame_bytes
 return frame_stored
num_padding_frames = int(padding_duration_ms / frame_duration_ms)
# use deque for the sliding window
ring_buffer = deque(maxlen=num_padding_frames)
# we have two states TRIGGERED and NOTTRIGGERED state
triggered = True #NOTTRIGGERED state

frames = frame_generator(buffer, frame_bytes)

speech_frame = []
for frame in frames:
    is_speech = vad.is_speech(frame, sample_rate)

Here is the error message:

TypeError Traceback (most recent call last) in 16 speech_frame = [] 17 for frame in frames: ---> 18 is_speech = vad.is_speech(frame, sample_rate) 19 #print(frames)

C:\Program Files\Python38\lib\site-packages\webrtcvad.py in is_speech(self, buf, sample_rate, length) 20 21 def is_speech(self, buf, sample_rate, length=None): ---> 22 length = length or int(len(buf) / 2) 23 if length * 2 > len(buf): 24 raise IndexError(

TypeError: object of type 'int' has no len()

Answer 1

I have solved it, you know vad.is_speech(buf=frame, sample_rate), it takes the buf and calculates it length, but an integer value does not posses the len() attributes in python. This throws an error for example:

num = 1
print(len(num))

Use this instead:

data = [1,2,3,4]
print(len(data))

So here is the correction to the code below:

frame_duration_ms=10
duration_in_ms = (frame_duration_ms / 1000) #duration in 10ms
frame_size = int(sample_rate * duration_in_ms) #frame size of 160
frame_bytes = frame_size * 2

values = []

def frame_generator(buffer, frame_bytes):
    # repeatedly store 320 length array to the frame_stored when the frame_bytes is less than the size of the buffer
    while offset+frame_bytes < len(buffer):
        frame_stored = buffer[offset : offset+frame_bytes]
        offset = offset + frame_bytes
        values.append(frame_stored)
 return values
num_padding_frames = int(padding_duration_ms / frame_duration_ms)
# use deque for the sliding window
ring_buffer = deque(maxlen=num_padding_frames)
# we have two states TRIGGERED and NOTTRIGGERED state
triggered = True #NOTTRIGGERED state

frames = frame_generator(buffer, frame_bytes)

frame = []
for frame in frames:
    is_speech = vad.is_speech(frame, sample_rate)