Enable redaction by default

This commit is contained in:
James Betker 2022-05-03 21:21:52 -06:00
parent c1d004aeb0
commit ddb19f6b0f
2 changed files with 86 additions and 32 deletions

View File

@ -165,7 +165,7 @@ class TextToSpeech:
Main entry point into Tortoise. Main entry point into Tortoise.
""" """
def __init__(self, autoregressive_batch_size=16, models_dir='.models', enable_redaction=False): def __init__(self, autoregressive_batch_size=16, models_dir='.models', enable_redaction=True):
""" """
Constructor Constructor
:param autoregressive_batch_size: Specifies how many samples to generate per batch. Lower this if you are seeing :param autoregressive_batch_size: Specifies how many samples to generate per batch. Lower this if you are seeing
@ -275,7 +275,6 @@ class TextToSpeech:
""" """
# Use generally found best tuning knobs for generation. # Use generally found best tuning knobs for generation.
kwargs.update({'temperature': .8, 'length_penalty': 1.0, 'repetition_penalty': 2.0, kwargs.update({'temperature': .8, 'length_penalty': 1.0, 'repetition_penalty': 2.0,
#'typical_sampling': True,
'top_p': .8, 'top_p': .8,
'cond_free_k': 2.0, 'diffusion_temperature': 1.0}) 'cond_free_k': 2.0, 'diffusion_temperature': 1.0})
# Presets are defined here. # Presets are defined here.

View File

@ -7,13 +7,52 @@ from transformers import Wav2Vec2ForCTC, Wav2Vec2FeatureExtractor, Wav2Vec2CTCTo
from tortoise.utils.audio import load_audio from tortoise.utils.audio import load_audio
def max_alignment(s1, s2, skip_character='~', record={}):
"""
A clever function that aligns s1 to s2 as best it can. Wherever a character from s1 is not found in s2, a '~' is
used to replace that character.
Finally got to use my DP skills!
"""
assert skip_character not in s1, f"Found the skip character {skip_character} in the provided string, {s1}"
if len(s1) == 0:
return ''
if len(s2) == 0:
return skip_character * len(s1)
if s1 == s2:
return s1
if s1[0] == s2[0]:
return s1[0] + max_alignment(s1[1:], s2[1:], skip_character, record)
take_s1_key = (len(s1), len(s2) - 1)
if take_s1_key in record:
take_s1, take_s1_score = record[take_s1_key]
else:
take_s1 = max_alignment(s1, s2[1:], skip_character, record)
take_s1_score = len(take_s1.replace(skip_character, ''))
record[take_s1_key] = (take_s1, take_s1_score)
take_s2_key = (len(s1) - 1, len(s2))
if take_s2_key in record:
take_s2, take_s2_score = record[take_s2_key]
else:
take_s2 = max_alignment(s1[1:], s2, skip_character, record)
take_s2_score = len(take_s2.replace(skip_character, ''))
record[take_s2_key] = (take_s2, take_s2_score)
return take_s1 if take_s1_score > take_s2_score else skip_character + take_s2
class Wav2VecAlignment: class Wav2VecAlignment:
"""
Uses wav2vec2 to perform audio<->text alignment.
"""
def __init__(self): def __init__(self):
self.model = Wav2Vec2ForCTC.from_pretrained("jbetker/wav2vec2-large-robust-ft-libritts-voxpopuli").cpu() self.model = Wav2Vec2ForCTC.from_pretrained("jbetker/wav2vec2-large-robust-ft-libritts-voxpopuli").cpu()
self.feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(f"facebook/wav2vec2-large-960h") self.feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(f"facebook/wav2vec2-large-960h")
self.tokenizer = Wav2Vec2CTCTokenizer.from_pretrained('jbetker/tacotron_symbols') self.tokenizer = Wav2Vec2CTCTokenizer.from_pretrained('jbetker/tacotron_symbols')
def align(self, audio, expected_text, audio_sample_rate=24000, topk=3, return_partial=False): def align(self, audio, expected_text, audio_sample_rate=24000):
orig_len = audio.shape[-1] orig_len = audio.shape[-1]
with torch.no_grad(): with torch.no_grad():
@ -25,32 +64,59 @@ class Wav2VecAlignment:
self.model = self.model.cpu() self.model = self.model.cpu()
logits = logits[0] logits = logits[0]
pred_string = self.tokenizer.decode(logits.argmax(-1).tolist())
fixed_expectation = max_alignment(expected_text, pred_string)
w2v_compression = orig_len // logits.shape[0] w2v_compression = orig_len // logits.shape[0]
expected_tokens = self.tokenizer.encode(expected_text) expected_tokens = self.tokenizer.encode(fixed_expectation)
expected_chars = list(fixed_expectation)
if len(expected_tokens) == 1: if len(expected_tokens) == 1:
return [0] # The alignment is simple; there is only one token. return [0] # The alignment is simple; there is only one token.
expected_tokens.pop(0) # The first token is a given. expected_tokens.pop(0) # The first token is a given.
next_expected_token = expected_tokens.pop(0) expected_chars.pop(0)
alignments = [0] alignments = [0]
def pop_till_you_win():
if len(expected_tokens) == 0:
return None
popped = expected_tokens.pop(0)
popped_char = expected_chars.pop(0)
while popped_char == '~':
alignments.append(-1)
if len(expected_tokens) == 0:
return None
popped = expected_tokens.pop(0)
popped_char = expected_chars.pop(0)
return popped
next_expected_token = pop_till_you_win()
for i, logit in enumerate(logits): for i, logit in enumerate(logits):
top = logit.topk(topk).indices.tolist() top = logit.argmax()
if next_expected_token in top: if next_expected_token == top:
alignments.append(i * w2v_compression) alignments.append(i * w2v_compression)
if len(expected_tokens) > 0: if len(expected_tokens) > 0:
next_expected_token = expected_tokens.pop(0) next_expected_token = pop_till_you_win()
else: else:
break break
if len(expected_tokens) > 0: pop_till_you_win()
print(f"Alignment did not work. {len(expected_tokens)} were not found, with the following string un-aligned:" assert len(expected_tokens) == 0, "This shouldn't happen. My coding sucks."
f" `{self.tokenizer.decode(expected_tokens)}`. Here's what wav2vec thought it heard:"
f"`{self.tokenizer.decode(logits.argmax(-1).tolist())}`")
if not return_partial:
return None
return alignments # Now fix up alignments. Anything with -1 should be interpolated.
alignments.append(orig_len) # This'll get removed but makes the algorithm below more readable.
for i in range(len(alignments)):
if alignments[i] == -1:
for j in range(i+1, len(alignments)):
if alignments[j] != -1:
next_found_token = j
break
for j in range(i, next_found_token):
gap = alignments[next_found_token] - alignments[i-1]
alignments[j] = (j-i+1) * gap // (next_found_token-i+1) + alignments[i-1]
def redact(self, audio, expected_text, audio_sample_rate=24000, topk=3): return alignments[:-1]
def redact(self, audio, expected_text, audio_sample_rate=24000):
if '[' not in expected_text: if '[' not in expected_text:
return audio return audio
splitted = expected_text.split('[') splitted = expected_text.split('[')
@ -58,33 +124,22 @@ class Wav2VecAlignment:
for spl in splitted[1:]: for spl in splitted[1:]:
assert ']' in spl, 'Every "[" character must be paired with a "]" with no nesting.' assert ']' in spl, 'Every "[" character must be paired with a "]" with no nesting.'
fully_split.extend(spl.split(']')) fully_split.extend(spl.split(']'))
# Remove any non-alphabetic character in the input text. This makes matching more likely.
fully_split = [re.sub(r'[^a-zA-Z ]', '', s) for s in fully_split]
# At this point, fully_split is a list of strings, with every other string being something that should be redacted. # At this point, fully_split is a list of strings, with every other string being something that should be redacted.
non_redacted_intervals = [] non_redacted_intervals = []
last_point = 0 last_point = 0
for i in range(len(fully_split)): for i in range(len(fully_split)):
if i % 2 == 0: if i % 2 == 0:
non_redacted_intervals.append((last_point, last_point + len(fully_split[i]) - 1)) end_interval = max(0, last_point + len(fully_split[i]) - 1)
non_redacted_intervals.append((last_point, end_interval))
last_point += len(fully_split[i]) last_point += len(fully_split[i])
bare_text = ''.join(fully_split) bare_text = ''.join(fully_split)
alignments = self.align(audio, bare_text, audio_sample_rate, topk, return_partial=True) alignments = self.align(audio, bare_text, audio_sample_rate)
# If alignment fails, we will attempt to recover by assuming the remaining alignments consume the rest of the string.
def get_alignment(i):
if i >= len(alignments):
return audio.shape[-1]
output_audio = [] output_audio = []
for nri in non_redacted_intervals: for nri in non_redacted_intervals:
start, stop = nri start, stop = nri
output_audio.append(audio[:, get_alignment(start):get_alignment(stop)]) output_audio.append(audio[:, alignments[start]:alignments[stop]])
return torch.cat(output_audio, dim=-1) return torch.cat(output_audio, dim=-1)
if __name__ == '__main__':
some_audio = load_audio('../../results/train_dotrice_0.wav', 24000)
aligner = Wav2VecAlignment()
text = "[God fucking damn it I'm so angry] The expressiveness of autoregressive transformers is literally nuts! I absolutely adore them."
redact = aligner.redact(some_audio, text)
torchaudio.save(f'test_output.wav', redact, 24000)