Spaces:
Running
Running
File size: 6,704 Bytes
1f4e6d7 |
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 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 |
import librosa
import torch
import torchaudio
class Slicer:
def __init__(self,
sr: int,
threshold: float = -40.,
min_length: int = 5000,
min_interval: int = 300,
hop_size: int = 20,
max_sil_kept: int = 5000):
if not min_length >= min_interval >= hop_size:
raise ValueError('The following condition must be satisfied: min_length >= min_interval >= hop_size')
if not max_sil_kept >= hop_size:
raise ValueError('The following condition must be satisfied: max_sil_kept >= hop_size')
min_interval = sr * min_interval / 1000
self.threshold = 10 ** (threshold / 20.)
self.hop_size = round(sr * hop_size / 1000)
self.win_size = min(round(min_interval), 4 * self.hop_size)
self.min_length = round(sr * min_length / 1000 / self.hop_size)
self.min_interval = round(min_interval / self.hop_size)
self.max_sil_kept = round(sr * max_sil_kept / 1000 / self.hop_size)
def _apply_slice(self, waveform, begin, end):
if len(waveform.shape) > 1:
return waveform[:, begin * self.hop_size: min(waveform.shape[1], end * self.hop_size)]
else:
return waveform[begin * self.hop_size: min(waveform.shape[0], end * self.hop_size)]
# @timeit
def slice(self, waveform):
if len(waveform.shape) > 1:
samples = librosa.to_mono(waveform)
else:
samples = waveform
if samples.shape[0] <= self.min_length:
return {"0": {"slice": False, "split_time": f"0,{len(waveform)}"}}
rms_list = librosa.feature.rms(y=samples, frame_length=self.win_size, hop_length=self.hop_size).squeeze(0)
sil_tags = []
silence_start = None
clip_start = 0
for i, rms in enumerate(rms_list):
# Keep looping while frame is silent.
if rms < self.threshold:
# Record start of silent frames.
if silence_start is None:
silence_start = i
continue
# Keep looping while frame is not silent and silence start has not been recorded.
if silence_start is None:
continue
# Clear recorded silence start if interval is not enough or clip is too short
is_leading_silence = silence_start == 0 and i > self.max_sil_kept
need_slice_middle = i - silence_start >= self.min_interval and i - clip_start >= self.min_length
if not is_leading_silence and not need_slice_middle:
silence_start = None
continue
# Need slicing. Record the range of silent frames to be removed.
if i - silence_start <= self.max_sil_kept:
pos = rms_list[silence_start: i + 1].argmin() + silence_start
if silence_start == 0:
sil_tags.append((0, pos))
else:
sil_tags.append((pos, pos))
clip_start = pos
elif i - silence_start <= self.max_sil_kept * 2:
pos = rms_list[i - self.max_sil_kept: silence_start + self.max_sil_kept + 1].argmin()
pos += i - self.max_sil_kept
pos_l = rms_list[silence_start: silence_start + self.max_sil_kept + 1].argmin() + silence_start
pos_r = rms_list[i - self.max_sil_kept: i + 1].argmin() + i - self.max_sil_kept
if silence_start == 0:
sil_tags.append((0, pos_r))
clip_start = pos_r
else:
sil_tags.append((min(pos_l, pos), max(pos_r, pos)))
clip_start = max(pos_r, pos)
else:
pos_l = rms_list[silence_start: silence_start + self.max_sil_kept + 1].argmin() + silence_start
pos_r = rms_list[i - self.max_sil_kept: i + 1].argmin() + i - self.max_sil_kept
if silence_start == 0:
sil_tags.append((0, pos_r))
else:
sil_tags.append((pos_l, pos_r))
clip_start = pos_r
silence_start = None
# Deal with trailing silence.
total_frames = rms_list.shape[0]
if silence_start is not None and total_frames - silence_start >= self.min_interval:
silence_end = min(total_frames, silence_start + self.max_sil_kept)
pos = rms_list[silence_start: silence_end + 1].argmin() + silence_start
sil_tags.append((pos, total_frames + 1))
# Apply and return slices.
if len(sil_tags) == 0:
return {"0": {"slice": False, "split_time": f"0,{len(waveform)}"}}
else:
chunks = []
# 第一段静音并非从头开始，补上有声片段
if sil_tags[0][0]:
chunks.append(
{"slice": False, "split_time": f"0,{min(waveform.shape[0], sil_tags[0][0] * self.hop_size)}"})
for i in range(0, len(sil_tags)):
# 标识有声片段（跳过第一段）
if i:
chunks.append({"slice": False,
"split_time": f"{sil_tags[i - 1][1] * self.hop_size},{min(waveform.shape[0], sil_tags[i][0] * self.hop_size)}"})
# 标识所有静音片段
chunks.append({"slice": True,
"split_time": f"{sil_tags[i][0] * self.hop_size},{min(waveform.shape[0], sil_tags[i][1] * self.hop_size)}"})
# 最后一段静音并非结尾，补上结尾片段
if sil_tags[-1][1] * self.hop_size < len(waveform):
chunks.append({"slice": False, "split_time": f"{sil_tags[-1][1] * self.hop_size},{len(waveform)}"})
chunk_dict = {}
for i in range(len(chunks)):
chunk_dict[str(i)] = chunks[i]
return chunk_dict
def cut(audio_path, db_thresh=-30, min_len=5000):
audio, sr = librosa.load(audio_path, sr=None)
slicer = Slicer(
sr=sr,
threshold=db_thresh,
min_length=min_len
)
chunks = slicer.slice(audio)
return chunks
def chunks2audio(audio_path, chunks):
chunks = dict(chunks)
audio, sr = torchaudio.load(audio_path)
if len(audio.shape) == 2 and audio.shape[1] >= 2:
audio = torch.mean(audio, dim=0).unsqueeze(0)
audio = audio.cpu().numpy()[0]
result = []
for k, v in chunks.items():
tag = v["split_time"].split(",")
if tag[0] != tag[1]:
result.append((v["slice"], audio[int(tag[0]):int(tag[1])]))
return result, sr
|