def main():
args = parser.parse_args()
sample = dict()
net_input = dict()
feature = get_feature(args.wav_path)
target_dict = Dictionary.load(args.target_dict_path)
model = load_model(args.w2v_path, target_dict)
model[0].eval()
generator = W2lViterbiDecoder(target_dict)
net_input["source"] = feature.unsqueeze(0)
padding_mask = torch.BoolTensor(
net_input["source"].size(1)).fill_(False).unsqueeze(0)
net_input["padding_mask"] = padding_mask
sample["net_input"] = net_input
with torch.no_grad():
hypo = generator.generate(model, sample, prefix_tokens=None)
hyp_pieces = target_dict.string(hypo[0][0]["tokens"].int().cpu())
print(post_process(hyp_pieces, 'letter'))
def predict_file(self, file_path):
generator = W2lViterbiDecoder(self.target_dict)
sample = dict()
net_input = dict()
feature = get_feature(file_path)
net_input["source"] = feature.unsqueeze(0).to(device)
padding_mask = torch.BoolTensor(
net_input["source"].size(1)).fill_(False).unsqueeze(0).to(device)
net_input["padding_mask"] = padding_mask
sample["net_input"] = net_input
with torch.no_grad():
hypo = generator.generate(self.model, sample, prefix_tokens=None)
# print(hypo[0][0]["tokens"].size())
hyp_pieces = self.target_dict.string(hypo[0][0]["tokens"].int().cpu(),
bpe_symbol='none')
asr_result = post_process(hyp_pieces, 'none')
# print(asr_result)
audio_embedding = self.model(**net_input)
audio_embedding = audio_embedding['encoder_out_no_proj'].squeeze(
1).cpu().numpy()
# audio_embedding = audio_embedding['encoder_out'].squeeze(1).detach().cpu().numpy()
# print(np.shape(audio_embedding))
return asr_result, audio_embedding
def build_generator(cfg: UnsupGenerateConfig):
w2l_decoder = cfg.w2l_decoder
if w2l_decoder == DecoderType.VITERBI:
from examples.speech_recognition.w2l_decoder import W2lViterbiDecoder
return W2lViterbiDecoder(cfg, task.target_dictionary)
elif w2l_decoder == DecoderType.KENLM:
from examples.speech_recognition.w2l_decoder import W2lKenLMDecoder
return W2lKenLMDecoder(cfg, task.target_dictionary)
elif w2l_decoder == DecoderType.FAIRSEQ:
from examples.speech_recognition.w2l_decoder import W2lFairseqLMDecoder
return W2lFairseqLMDecoder(cfg, task.target_dictionary)
elif w2l_decoder == DecoderType.KALDI:
from examples.speech_recognition.kaldi.kaldi_decoder import KaldiDecoder
assert cfg.kaldi_decoder_config is not None
return KaldiDecoder(
cfg.kaldi_decoder_config,
cfg.beam,
)
else:
raise NotImplementedError(
"only wav2letter decoders with (viterbi, kenlm, fairseqlm) options are supported at the moment but found "
+ str(w2l_decoder)
)
def build_generator(
self, models, args, seq_gen_cls=None, extra_gen_cls_kwargs=None
):
if self.args.decoder_type == "ctc":
from examples.speech_recognition.w2l_decoder import W2lViterbiDecoder
return W2lViterbiDecoder(args, self.tgt_dict)
else:
raise NotImplementedError("only ctc decoder is supported at the moment")
def build_generator(self, task, models, cfg):
try:
from examples.speech_recognition.w2l_decoder import W2lViterbiDecoder
except Exception:
raise Exception(
"Cannot run this test without flashlight dependency")
with open_dict(cfg):
cfg.nbest = 1
return W2lViterbiDecoder(cfg, task.target_dictionary)
def build_generator(self, args):
w2l_decoder = getattr(args, "w2l_decoder", None)
if w2l_decoder == "viterbi":
from examples.speech_recognition.w2l_decoder import W2lViterbiDecoder
return W2lViterbiDecoder(args, self.target_dictionary)
elif w2l_decoder == "kenlm":
from examples.speech_recognition.w2l_decoder import W2lKenLMDecoder
return W2lKenLMDecoder(args, self.target_dictionary)
else:
return super().build_generator(args)
def _init_model(self, dict_file):
parser = self._create_parser(dict_file)
model_dir = os.path.dirname(self.model_path)
target_dict_path = os.path.join(model_dir, dict_file)
args = parser.parse_args([
'--target_dict_path', target_dict_path, '--w2v_path',
self.model_path
])
target_dict = Dictionary.load(args.target_dict_path)
self.model = self._load_model(args.w2v_path, target_dict)[0]
self.model.eval()
self.generator = W2lViterbiDecoder(args, target_dict)
self.args = args
self.target_dict = target_dict
def get_decoder(decoder_args_dict, dictionary):
decoder_args = Namespace(**decoder_args_dict)
if decoder_args.decoder_type == "viterbi":
from examples.speech_recognition.w2l_decoder import W2lViterbiDecoder
return W2lViterbiDecoder(decoder_args, dictionary)
elif decoder_args.decoder_type == "kenlm":
from examples.speech_recognition.w2l_decoder import W2lKenLMDecoder
decoder_args.beam_size_token = len(dictionary)
if isinstance(decoder_args.unk_weight, str):
decoder_args.unk_weight = eval(decoder_args.unk_weight)
return W2lKenLMDecoder(decoder_args, dictionary)
return None
def build_generator(args):
w2l_decoder = getattr(args, "w2l_decoder", None)
if w2l_decoder == "viterbi":
from examples.speech_recognition.w2l_decoder import W2lViterbiDecoder
return W2lViterbiDecoder(args, task.target_dictionary)
elif w2l_decoder == "kenlm":
from examples.speech_recognition.w2l_decoder import W2lKenLMDecoder
print(task.target_dictionary.symbols)
return W2lKenLMDecoder(args, task.target_dictionary)
elif w2l_decoder == "fairseqlm":
from examples.speech_recognition.w2l_decoder import W2lFairseqLMDecoder
print(task.target_dictionary.symbols)
return W2lFairseqLMDecoder(args, task.target_dictionary)
else:
return super().build_generator(args)
def build_generator(args):
w2l_decoder = getattr(args, "w2l_decoder", None)
if w2l_decoder == "viterbi":
from examples.speech_recognition.w2l_decoder import W2lViterbiDecoder
return W2lViterbiDecoder(args, task.target_dictionary)
elif w2l_decoder == "kenlm":
from examples.speech_recognition.w2l_decoder import W2lKenLMDecoder
return W2lKenLMDecoder(args, task.target_dictionary)
elif w2l_decoder == "fairseqlm":
from examples.speech_recognition.w2l_decoder import W2lFairseqLMDecoder
return W2lFairseqLMDecoder(args, task.target_dictionary)
else:
print('only wav2letter decoders with (viterbi, kenlm, fairseqlm) options are supported at the moment')
def main():
parser = argparse.ArgumentParser(description='Wav2vec-2.0 Recognize')
parser.add_argument('--wav_path',
type=str,
default='~/xxx.wav',
help='path of wave file')
parser.add_argument('--w2v_path',
type=str,
default='pre_train_weights/wav2vec_vox_960h_pl.pt',
help='path of pre-trained wav2vec-2.0 model')
parser.add_argument('--target_dict_path',
type=str,
default='pre_train_weights/dict.ltr.txt',
help='path of target dict (dict.ltr.txt)')
args = parser.parse_args()
sample = dict()
net_input = dict()
feature = get_feature(args.wav_path)
target_dict = Dictionary.load(args.target_dict_path)
model = load_model(args.w2v_path, target_dict)
model.eval()
generator = W2lViterbiDecoder(target_dict)
net_input["source"] = feature.unsqueeze(0).to(device)
padding_mask = torch.BoolTensor(
net_input["source"].size(1)).fill_(False).unsqueeze(0).to(device)
net_input["padding_mask"] = padding_mask
sample["net_input"] = net_input
with torch.no_grad():
hypo = generator.generate(model, sample, prefix_tokens=None)
print(hypo[0][0]["tokens"].size())
hyp_pieces = target_dict.string(hypo[0][0]["tokens"].int().cpu(),
bpe_symbol='none')
print(len(hyp_pieces.split()))
print(post_process(hyp_pieces, 'letter_b'))
def build_generator(args):
w2l_decoder = getattr(args, "w2l_decoder", None)
if w2l_decoder == "viterbi":
from examples.speech_recognition.w2l_decoder import W2lViterbiDecoder
return W2lViterbiDecoder(args, task.target_dictionary)
elif w2l_decoder == "kenlm":
from examples.speech_recognition.w2l_decoder import W2lKenLMDecoder
return W2lKenLMDecoder(args, task.target_dictionary)
elif w2l_decoder == "fairseqlm":
from examples.speech_recognition.w2l_decoder import W2lFairseqLMDecoder
return W2lFairseqLMDecoder(args, task.target_dictionary)
elif w2l_decoder == "ctc_decoder":
from examples.speech_recognition.ctc_decoder import CTCDecoder
return CTCDecoder(args, task.target_dictionary)
elif w2l_decoder == "cif_decoder":
from examples.speech_recognition.cif_decoder import CIFDecoder
return CIFDecoder(args, task.target_dictionary, {})
elif w2l_decoder == "cif_lm_decoder":
from examples.speech_recognition.cif_decoder import CIFDecoder
return CIFDecoder(args, task.target_dictionary, ({}, {}))
elif w2l_decoder == "cif_bert_decoder":
from examples.speech_recognition.cif_bert_decoder import CIF_BERT_Decoder
return CIF_BERT_Decoder(args, task.target_dictionary)
elif w2l_decoder == "seq2seq_decoder":
from examples.speech_recognition.seq2seq_decoder import Seq2seqDecoder
return Seq2seqDecoder(args, task.target_dictionary, {})
elif w2l_decoder == "seq2seq_lm_decoder":
from examples.speech_recognition.seq2seq_decoder import Seq2seqDecoder
return Seq2seqDecoder(args, task.target_dictionary, ({}, {}))
else:
return super().build_generator(args)
def __init__(self, model_weight, target_dict):
self.target_dict = Dictionary.load(target_dict)
self.model = load_model(model_weight, self.target_dict)
self.model.eval()
self.generator = W2lViterbiDecoder(self.target_dict)
def _main(cfg: DictConfig, output_file):
logging.basicConfig(
format="%(asctime)s | %(levelname)s | %(name)s | %(message)s",
datefmt="%Y-%m-%d %H:%M:%S",
level=os.environ.get("LOGLEVEL", "INFO").upper(),
stream=output_file,
)
if 'label_dir' in cfg.task:
manifest_dir, _ = os.path.split(cfg.dataset.gen_subset)
with read_write(cfg):
cfg.task.label_dir = os.path.join(cfg.task.data, manifest_dir)
print('cfg.task.data', cfg.task.label_dir)
logger = logging.getLogger("fairseq_cli.generate")
utils.import_user_module(cfg.common)
if cfg.dataset.max_tokens is None and cfg.dataset.batch_size is None:
cfg.dataset.max_tokens = 12000
logger.info(cfg)
# Fix seed for stochastic decoding
if cfg.common.seed is not None and not cfg.generation.no_seed_provided:
np.random.seed(cfg.common.seed)
utils.set_torch_seed(cfg.common.seed)
use_cuda = torch.cuda.is_available() and not cfg.common.cpu
# Load dataset splits
task = tasks.setup_task(cfg.task)
# Set dictionaries
try:
src_dict = getattr(task, "source_dictionary", None)
except NotImplementedError:
src_dict = None
tgt_dict = task.target_dictionary
overrides = ast.literal_eval(cfg.common_eval.model_overrides)
# Load ensemble
logger.info("loading model(s) from {}".format(cfg.common_eval.path))
models, saved_cfg = checkpoint_utils.load_model_ensemble(
utils.split_paths(cfg.common_eval.path),
arg_overrides=overrides,
task=task,
suffix=cfg.checkpoint.checkpoint_suffix,
strict=(cfg.checkpoint.checkpoint_shard_count == 1),
num_shards=cfg.checkpoint.checkpoint_shard_count,
)
token_type = None
if type(models[0]) == Wav2Bart or type(models[0]) == WavTransBart or type(models[0]) == WavLinearBart or type(models[0]) == WavBart2Bart:
token_type = 'bart'
elif type(models[0]) == Wav2BartChr:
token_type = 'chr'
elif type(models[0]) == Wav2VecCtc or type(models[0]) == Wav2BertChr or type(models[0]) == Wav2BertMixChr:
token_type = 'chrctc'
elif type(models[0]) == Wav2Bert:
token_type = 'bert'
else:
raise ValueError(f'token_type not defined for {type(models[0])}')
print(f'token_type is {token_type}')
# loading the dataset should happen after the checkpoint has been loaded so we can give it the saved task config
task.load_dataset(cfg.dataset.gen_subset, task_cfg=saved_cfg.task)
if cfg.generation.lm_path is not None:
overrides["data"] = cfg.task.data
try:
lms, _ = checkpoint_utils.load_model_ensemble(
[cfg.generation.lm_path], arg_overrides=overrides, task=None
)
except:
logger.warning(
f"Failed to load language model! Please make sure that the language model dict is the same "
f"as target dict and is located in the data dir ({cfg.task.data})"
)
raise
assert len(lms) == 1
else:
lms = [None]
# Optimize ensemble for generation
for model in chain(models, lms):
if model is None:
continue
if cfg.common.fp16:
model.half()
if use_cuda and not cfg.distributed_training.pipeline_model_parallel:
model.cuda()
model.prepare_for_inference_(cfg)
# Load alignment dictionary for unknown word replacement
# (None if no unknown word replacement, empty if no path to align dictionary)
align_dict = utils.load_align_dict(cfg.generation.replace_unk)
itr = task.get_batch_iterator(
dataset=task.dataset(cfg.dataset.gen_subset),
max_tokens=cfg.dataset.max_tokens,
max_sentences=cfg.dataset.batch_size,
max_positions=utils.resolve_max_positions(
task.max_positions(), *[m.max_positions() for m in models]
),
ignore_invalid_inputs=cfg.dataset.skip_invalid_size_inputs_valid_test,
required_batch_size_multiple=cfg.dataset.required_batch_size_multiple,
seed=cfg.common.seed,
num_shards=cfg.distributed_training.distributed_world_size,
shard_id=cfg.distributed_training.distributed_rank,
num_workers=cfg.dataset.num_workers,
data_buffer_size=cfg.dataset.data_buffer_size,
).next_epoch_itr(shuffle=False)
progress = progress_bar.progress_bar(
itr,
log_format=cfg.common.log_format,
log_interval=cfg.common.log_interval,
default_log_format=("tqdm" if not cfg.common.no_progress_bar else "simple"),
)
# Initialize generator
gen_timer = StopwatchMeter()
extra_gen_cls_kwargs = {"lm_model": lms[0], "lm_weight": cfg.generation.lm_weight}
print('cfg.generation', cfg.generation)
# print(cfg.task._name == 'audio_pretraining')
if cfg.task._name != 'audio_pretraining' and cfg.task._name != 'audio_pretraining_bertbpe':
generator = task.build_generator(
models, cfg.generation, extra_gen_cls_kwargs=extra_gen_cls_kwargs
)
else:
print('use W2lViterbiDecoder')
from examples.speech_recognition.w2l_decoder import W2lViterbiDecoder
from easydict import EasyDict as edict
args = edict({
'criterion': 'ctc',
'nbest': 1,
})
generator = W2lViterbiDecoder(args, task.target_dictionary)
# Handle tokenization and BPE
tokenizer = task.build_tokenizer(cfg.tokenizer)
bpe = task.build_bpe(cfg.bpe)
def decode_fn(x):
if bpe is not None:
x = bpe.decode(x)
if tokenizer is not None:
x = tokenizer.decode(x)
return x
scorer = scoring.build_scorer(cfg.scoring, tgt_dict)
num_sentences = 0
has_target = True
wps_meter = TimeMeter()
for si, sample in enumerate(progress):
sample = utils.move_to_cuda(sample) if use_cuda else sample
if "net_input" not in sample:
continue
prefix_tokens = None
if cfg.generation.prefix_size > 0:
prefix_tokens = sample["target"][:, : cfg.generation.prefix_size]
constraints = None
if "constraints" in sample:
constraints = sample["constraints"]
gen_timer.start()
hypos = task.inference_step(
generator,
models,
sample,
prefix_tokens=prefix_tokens,
constraints=constraints,
)
# print('hypos', hypos)
num_generated_tokens = sum(len(h[0]["tokens"]) for h in hypos)
gen_timer.stop(num_generated_tokens)
for i, sample_id in enumerate(sample["id"].tolist()):
has_target = sample["target"] is not None
# Remove padding
if "src_tokens" in sample["net_input"]:
src_tokens = utils.strip_pad(
sample["net_input"]["src_tokens"][i, :], tgt_dict.pad()
)
else:
src_tokens = None
target_tokens = None
if has_target:
target_tokens = (
utils.strip_pad(sample["target"][i, :], tgt_dict.pad()).int().cpu()
)
# Either retrieve the original sentences or regenerate them from tokens.
if align_dict is not None:
src_str = task.dataset(cfg.dataset.gen_subset).src.get_original_text(
sample_id
)
target_str = task.dataset(cfg.dataset.gen_subset).tgt.get_original_text(
sample_id
)
else:
if src_dict is not None:
src_str = src_dict.string(src_tokens, cfg.common_eval.post_process)
else:
src_str = ""
if has_target:
if token_type == 'chr':
target_str = tgt_dict.string(
target_tokens,
cfg.common_eval.post_process,
escape_unk=True,
extra_symbols_to_ignore=get_symbols_to_strip_from_output(
generator
),
)
elif token_type == 'bart':
target_str = task.bart.decode(target_tokens.int().cpu())
elif token_type == 'bert':
target_str = task.bert.decode(target_tokens.int().cpu())
elif token_type == 'chrctc':
target_str = tgt_dict.string(
target_tokens,
cfg.common_eval.post_process,
escape_unk=True,
)
else:
raise ValueError(f'token_type not defined for {type(models[0])}')
src_str = decode_fn(src_str)
if has_target and token_type == 'chr':
target_str = decode_fn(target_str)
elif has_target and token_type == 'chrctc':
target_str = ''.join(target_str.split()).replace('|', ' ')
if not cfg.common_eval.quiet:
if src_dict is not None:
print("S-{}\t{}".format(sample_id, src_str), file=output_file)
if has_target:
print("T-{}\t{}".format(sample_id, target_str), file=output_file)
# Process top predictions
for j, hypo in enumerate(hypos[i][: cfg.generation.nbest]):
# print('align', hypo["alignment"])
if token_type == 'bart':
hypo_tokens = hypo["tokens"].int().cpu()
hypo_str = task.bart.decode(hypo["tokens"].int().cpu())
alignment = hypo["alignment"]
elif token_type == 'chr':
hypo_tokens, hypo_str, alignment = utils.post_process_prediction(
hypo_tokens=hypo["tokens"].int().cpu(),
src_str=src_str,
alignment=hypo["alignment"],
align_dict=align_dict,
tgt_dict=tgt_dict,
remove_bpe=cfg.common_eval.post_process,
# extra_symbols_to_ignore=get_symbols_to_strip_from_output(generator),
)
elif token_type == 'chrctc':
hypo_tokens = hypo["tokens"].int().cpu()
hypo_str = task.target_dictionary.string(hypo_tokens)
hypo["positional_scores"] = torch.FloatTensor([0.])
elif token_type == 'bert':
hypo_tokens = hypo["tokens"].int().cpu()
hypo_str = task.bert.decode(hypo["tokens"].int().cpu())
alignment = hypo["alignment"]
else:
raise ValueError(f'token_type not defined for {type(models[0])}')
detok_hypo_str = decode_fn(hypo_str)
if token_type == 'chr' or token_type == 'chrctc':
print('target_str', ''.join(target_str.split()).replace('|', ' '))
print('typo_str', ''.join(detok_hypo_str.split()).replace('|', ' '))
detok_hypo_str = ''.join(detok_hypo_str.split()).replace('|', ' ')
# target_str = ''.join(target_str.split()).replace('|', ' ')
elif token_type == 'bart':
print('target_str', target_str)
print('typo_str', detok_hypo_str)
#elif token_type == 'chrctc':
# print('target_str', ''.join(target_str.split()).replace('|', ' '))
# print('typo_str', ''.join(detok_hypo_str.split()).replace('|', ' '))
if not cfg.common_eval.quiet:
score = hypo["score"] / math.log(2) # convert to base 2
# original hypothesis (after tokenization and BPE)
print(
"H-{}\t{}\t{}".format(sample_id, score, hypo_str),
file=output_file,
)
# detokenized hypothesis
print(
"D-{}\t{}\t{}".format(sample_id, score, detok_hypo_str),
file=output_file,
)
print(
"P-{}\t{}".format(
sample_id,
" ".join(
map(
lambda x: "{:.4f}".format(x),
# convert from base e to base 2
hypo["positional_scores"]
.div_(math.log(2))
.tolist(),
)
),
),
file=output_file,
)
if cfg.generation.print_alignment == "hard":
print(
"A-{}\t{}".format(
sample_id,
" ".join(
[
"{}-{}".format(src_idx, tgt_idx)
for src_idx, tgt_idx in alignment
]
),
),
file=output_file,
)
if cfg.generation.print_alignment == "soft":
print(
"A-{}\t{}".format(
sample_id,
" ".join(
[
",".join(src_probs)
for src_probs in alignment
]
),
),
file=output_file,
)
if cfg.generation.print_step:
print(
"I-{}\t{}".format(sample_id, hypo["steps"]),
file=output_file,
)
if cfg.generation.retain_iter_history:
for step, h in enumerate(hypo["history"]):
_, h_str, _ = utils.post_process_prediction(
hypo_tokens=h["tokens"].int().cpu(),
src_str=src_str,
alignment=None,
align_dict=None,
tgt_dict=tgt_dict,
remove_bpe=None,
)
print(
"E-{}_{}\t{}".format(sample_id, step, h_str),
file=output_file,
)
# Score only the top hypothesis
if has_target and j == 0:
if align_dict is not None or cfg.common_eval.post_process is not None:
# Convert back to tokens for evaluation with unk replacement and/or without BPE
target_tokens = tgt_dict.encode_line(
target_str, add_if_not_exist=True
)
hypo_tokens = tgt_dict.encode_line(
detok_hypo_str, add_if_not_exist=True
)
if hasattr(scorer, "add_string"):
# print('add_string 1', target_str, '2', detok_hypo_str)
# if si > 2:
# raise
print('2', target_str, detok_hypo_str)
scorer.add_string(target_str, detok_hypo_str)
else:
scorer.add(target_tokens, hypo_tokens)
wps_meter.update(num_generated_tokens)
progress.log({"wps": round(wps_meter.avg)})
num_sentences += (
sample["nsentences"] if "nsentences" in sample else sample["id"].numel()
)
logger.info("NOTE: hypothesis and token scores are output in base 2")
logger.info(
"Translated {:,} sentences ({:,} tokens) in {:.1f}s ({:.2f} sentences/s, {:.2f} tokens/s)".format(
num_sentences,
gen_timer.n,
gen_timer.sum,
num_sentences / gen_timer.sum,
1.0 / gen_timer.avg,
)
)
if has_target:
if cfg.bpe and not cfg.generation.sacrebleu:
if cfg.common_eval.post_process:
logger.warning(
"BLEU score is being computed by splitting detokenized string on spaces, this is probably not what you want. Use --sacrebleu for standard 13a BLEU tokenization"
)
else:
logger.warning(
"If you are using BPE on the target side, the BLEU score is computed on BPE tokens, not on proper words. Use --sacrebleu for standard 13a BLEU tokenization"
)
# use print to be consistent with other main outputs: S-, H-, T-, D- and so on
print(
"Generate {} with beam={}: {}".format(
cfg.dataset.gen_subset, cfg.generation.beam, scorer.result_string()
),
file=output_file,
)
return scorer