python/FunASR-XL.git

			@@ -17,9 +17,6 @@
			from distutils.version import LooseVersion

			from funasr.register import tables
			from funasr.losses.label_smoothing_loss import (
			LabelSmoothingLoss, # noqa: H301
			)
			from funasr.utils import postprocess_utils
			from funasr.metrics.compute_acc import th_accuracy
			from funasr.models.paraformer.model import Paraformer
			@@ -47,14 +44,14 @@
			FunASR: A Fundamental End-to-End Speech Recognition Toolkit
			https://arxiv.org/abs/2305.11013
			"""


			def __init__(
			self,
			*args,
			**kwargs,
			):
			super().__init__(args, *kwargs)


			self.target_buffer_length = kwargs.get("target_buffer_length", -1)
			inner_dim = kwargs.get("inner_dim", 256)
			bias_encoder_type = kwargs.get("bias_encoder_type", "lstm")
			@@ -63,15 +60,16 @@
			crit_attn_smooth = kwargs.get("crit_attn_smooth", 0.0)
			bias_encoder_dropout_rate = kwargs.get("bias_encoder_dropout_rate", 0.0)


			if bias_encoder_type == 'lstm':
			self.bias_encoder = torch.nn.LSTM(inner_dim, inner_dim, 1, batch_first=True, dropout=bias_encoder_dropout_rate)
			if bias_encoder_type == "lstm":
			self.bias_encoder = torch.nn.LSTM(
			inner_dim, inner_dim, 1, batch_first=True, dropout=bias_encoder_dropout_rate
			)
			self.bias_embed = torch.nn.Embedding(self.vocab_size, inner_dim)
			elif bias_encoder_type == 'mean':
			elif bias_encoder_type == "mean":
			self.bias_embed = torch.nn.Embedding(self.vocab_size, inner_dim)
			else:
			logging.error("Unsupport bias encoder type: {}".format(bias_encoder_type))


			if self.target_buffer_length > 0:
			self.hotword_buffer = None
			self.length_record = []
			@@ -82,7 +80,6 @@
			self.attn_loss = torch.nn.L1Loss()
			self.crit_attn_smooth = crit_attn_smooth


			def forward(
			self,
			speech: torch.Tensor,
			@@ -92,74 +89,73 @@
			**kwargs,
			) -> Tuple[torch.Tensor, Dict[str, torch.Tensor], torch.Tensor]:
			"""Frontend + Encoder + Decoder + Calc loss


			Args:
			speech: (Batch, Length, ...)
			speech_lengths: (Batch, )
			text: (Batch, Length)
			text_lengths: (Batch,)
			"""
			if len(text_lengths.size()) > 1:
			text_lengths = text_lengths[:, 0]
			if len(speech_lengths.size()) > 1:
			speech_lengths = speech_lengths[:, 0]

			text_lengths = text_lengths.squeeze()
			speech_lengths = speech_lengths.squeeze()

			batch_size = speech.shape[0]

			hotword_pad = kwargs.get("hotword_pad")
			hotword_lengths = kwargs.get("hotword_lengths")
			dha_pad = kwargs.get("dha_pad")

			# dha_pad = kwargs.get("dha_pad")

			# 1. Encoder
			encoder_out, encoder_out_lens = self.encode(speech, speech_lengths)


			loss_ctc, cer_ctc = None, None


			stats = dict()


			# 1. CTC branch
			if self.ctc_weight != 0.0:
			loss_ctc, cer_ctc = self._calc_ctc_loss(
			encoder_out, encoder_out_lens, text, text_lengths
			)


			# Collect CTC branch stats
			stats["loss_ctc"] = loss_ctc.detach() if loss_ctc is not None else None
			stats["cer_ctc"] = cer_ctc


			# 2b. Attention decoder branch
			loss_att, acc_att, cer_att, wer_att, loss_pre, loss_ideal = self._calc_att_clas_loss(
			encoder_out, encoder_out_lens, text, text_lengths, hotword_pad, hotword_lengths
			)


			# 3. CTC-Att loss definition
			if self.ctc_weight == 0.0:
			loss = loss_att + loss_pre * self.predictor_weight
			else:
			loss = self.ctc_weight * loss_ctc + (1 - self.ctc_weight) * loss_att + loss_pre * self.predictor_weight

			loss = (
			self.ctc_weight * loss_ctc
			+ (1 - self.ctc_weight) * loss_att
			+ loss_pre * self.predictor_weight
			)

			if loss_ideal is not None:
			loss = loss + loss_ideal * self.crit_attn_weight
			stats["loss_ideal"] = loss_ideal.detach().cpu()


			# Collect Attn branch stats
			stats["loss_att"] = loss_att.detach() if loss_att is not None else None
			stats["acc"] = acc_att
			stats["cer"] = cer_att
			stats["wer"] = wer_att
			stats["loss_pre"] = loss_pre.detach().cpu() if loss_pre is not None else None


			stats["loss"] = torch.clone(loss.detach())
			# force_gatherable: to-device and to-tensor if scalar for DataParallel
			if self.length_normalized_loss:
			batch_size = int((text_lengths + self.predictor_bias).sum())


			loss, stats, weight = force_gatherable((loss, stats, batch_size), loss.device)
			return loss, stats, weight



			def _calc_att_clas_loss(
			self,
			encoder_out: torch.Tensor,
			@@ -169,39 +165,53 @@
			hotword_pad: torch.Tensor,
			hotword_lengths: torch.Tensor,
			):
			encoder_out_mask = (~make_pad_mask(encoder_out_lens, maxlen=encoder_out.size(1))[:, None, :]).to(
			encoder_out.device)
			encoder_out_mask = (
			~make_pad_mask(encoder_out_lens, maxlen=encoder_out.size(1))[:, None, :]
			).to(encoder_out.device)

			if self.predictor_bias == 1:
			_, ys_pad = add_sos_eos(ys_pad, self.sos, self.eos, self.ignore_id)
			ys_pad_lens = ys_pad_lens + self.predictor_bias
			pre_acoustic_embeds, pre_token_length, _, _ = self.predictor(encoder_out, ys_pad, encoder_out_mask,
			ignore_id=self.ignore_id)


			pre_acoustic_embeds, pre_token_length, _, _ = self.predictor(
			encoder_out, ys_pad, encoder_out_mask, ignore_id=self.ignore_id
			)
			# -1. bias encoder
			if self.use_decoder_embedding:
			hw_embed = self.decoder.embed(hotword_pad)
			else:
			hw_embed = self.bias_embed(hotword_pad)

			hw_embed, (_, _) = self.bias_encoder(hw_embed)
			_ind = np.arange(0, hotword_pad.shape[0]).tolist()
			selected = hw_embed[_ind, [i - 1 for i in hotword_lengths.detach().cpu().tolist()]]
			contextual_info = selected.squeeze(0).repeat(ys_pad.shape[0], 1, 1).to(ys_pad.device)


			# 0. sampler
			decoder_out_1st = None
			if self.sampling_ratio > 0.0:

			sematic_embeds, decoder_out_1st = self.sampler(encoder_out, encoder_out_lens, ys_pad, ys_pad_lens,
			pre_acoustic_embeds, contextual_info)
			sematic_embeds, decoder_out_1st = self.sampler(
			encoder_out,
			encoder_out_lens,
			ys_pad,
			ys_pad_lens,
			pre_acoustic_embeds,
			contextual_info,
			)
			else:
			sematic_embeds = pre_acoustic_embeds


			# 1. Forward decoder
			decoder_outs = self.decoder(
			encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, contextual_info=contextual_info
			encoder_out,
			encoder_out_lens,
			sematic_embeds,
			ys_pad_lens,
			contextual_info=contextual_info,
			)
			decoder_out, _ = decoder_outs[0], decoder_outs[1]
			'''
			"""
			if self.crit_attn_weight > 0 and attn.shape[-1] > 1:
			ideal_attn = ideal_attn + self.crit_attn_smooth / (self.crit_attn_smooth + 1.0)
			attn_non_blank = attn[:,:,:,:-1]
			@@ -209,9 +219,9 @@
			loss_ideal = self.attn_loss(attn_non_blank.max(1)[0], ideal_attn_non_blank.to(attn.device))
			else:
			loss_ideal = None
			'''
			"""
			loss_ideal = None


			if decoder_out_1st is None:
			decoder_out_1st = decoder_out
			# 2. Compute attention loss
			@@ -222,19 +232,28 @@
			ignore_label=self.ignore_id,
			)
			loss_pre = self.criterion_pre(ys_pad_lens.type_as(pre_token_length), pre_token_length)


			# Compute cer/wer using attention-decoder
			if self.training or self.error_calculator is None:
			cer_att, wer_att = None, None
			else:
			ys_hat = decoder_out_1st.argmax(dim=-1)
			cer_att, wer_att = self.error_calculator(ys_hat.cpu(), ys_pad.cpu())


			return loss_att, acc_att, cer_att, wer_att, loss_pre, loss_ideal


			def sampler(self, encoder_out, encoder_out_lens, ys_pad, ys_pad_lens, pre_acoustic_embeds, contextual_info):
			tgt_mask = (~make_pad_mask(ys_pad_lens, maxlen=ys_pad_lens.max())[:, :, None]).to(ys_pad.device)

			def sampler(
			self,
			encoder_out,
			encoder_out_lens,
			ys_pad,
			ys_pad_lens,
			pre_acoustic_embeds,
			contextual_info,
			):
			tgt_mask = (~make_pad_mask(ys_pad_lens, maxlen=ys_pad_lens.max())[:, :, None]).to(
			ys_pad.device
			)
			ys_pad = ys_pad * tgt_mask[:, :, 0]
			if self.share_embedding:
			ys_pad_embed = self.decoder.output_layer.weight[ys_pad]
			@@ -242,7 +261,11 @@
			ys_pad_embed = self.decoder.embed(ys_pad)
			with torch.no_grad():
			decoder_outs = self.decoder(
			encoder_out, encoder_out_lens, pre_acoustic_embeds, ys_pad_lens, contextual_info=contextual_info
			encoder_out,
			encoder_out_lens,
			pre_acoustic_embeds,
			ys_pad_lens,
			contextual_info=contextual_info,
			)
			decoder_out, _ = decoder_outs[0], decoder_outs[1]
			pred_tokens = decoder_out.argmax(-1)
			@@ -252,22 +275,35 @@
			input_mask = torch.ones_like(nonpad_positions)
			bsz, seq_len = ys_pad.size()
			for li in range(bsz):
			target_num = (((seq_lens[li] - same_num[li].sum()).float()) * self.sampling_ratio).long()
			target_num = (
			((seq_lens[li] - same_num[li].sum()).float()) * self.sampling_ratio
			).long()
			if target_num > 0:
			input_mask[li].scatter_(dim=0,
			index=torch.randperm(seq_lens[li])[:target_num].to(pre_acoustic_embeds.device),
			value=0)
			input_mask[li].scatter_(
			dim=0,
			index=torch.randperm(seq_lens[li])[:target_num].to(
			pre_acoustic_embeds.device
			),
			value=0,
			)
			input_mask = input_mask.eq(1)
			input_mask = input_mask.masked_fill(~nonpad_positions, False)
			input_mask_expand_dim = input_mask.unsqueeze(2).to(pre_acoustic_embeds.device)

			sematic_embeds = pre_acoustic_embeds.masked_fill(~input_mask_expand_dim, 0) + ys_pad_embed.masked_fill(
			input_mask_expand_dim, 0)

			sematic_embeds = pre_acoustic_embeds.masked_fill(
			~input_mask_expand_dim, 0
			) + ys_pad_embed.masked_fill(input_mask_expand_dim, 0)
			return sematic_embeds * tgt_mask, decoder_out * tgt_mask


			def cal_decoder_with_predictor(self, encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, hw_list=None,
			clas_scale=1.0):

			def cal_decoder_with_predictor(
			self,
			encoder_out,
			encoder_out_lens,
			sematic_embeds,
			ys_pad_lens,
			hw_list=None,
			clas_scale=1.0,
			):
			if hw_list is None:
			hw_list = [torch.Tensor([1]).long().to(encoder_out.device)] # empty hotword list
			hw_list_pad = pad_list(hw_list, 0)
			@@ -279,101 +315,129 @@
			hw_embed = h_n.repeat(encoder_out.shape[0], 1, 1)
			else:
			hw_lengths = [len(i) for i in hw_list]
			hw_list_pad = pad_list([torch.Tensor(i).long() for i in hw_list], 0).to(encoder_out.device)
			hw_list_pad = pad_list([torch.Tensor(i).long() for i in hw_list], 0).to(
			encoder_out.device
			)
			if self.use_decoder_embedding:
			hw_embed = self.decoder.embed(hw_list_pad)
			else:
			hw_embed = self.bias_embed(hw_list_pad)
			hw_embed = torch.nn.utils.rnn.pack_padded_sequence(hw_embed, hw_lengths, batch_first=True,
			enforce_sorted=False)
			hw_embed = torch.nn.utils.rnn.pack_padded_sequence(
			hw_embed, hw_lengths, batch_first=True, enforce_sorted=False
			)
			_, (h_n, _) = self.bias_encoder(hw_embed)
			hw_embed = h_n.repeat(encoder_out.shape[0], 1, 1)


			decoder_outs = self.decoder(
			encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, contextual_info=hw_embed, clas_scale=clas_scale
			encoder_out,
			encoder_out_lens,
			sematic_embeds,
			ys_pad_lens,
			contextual_info=hw_embed,
			clas_scale=clas_scale,
			)

			decoder_out = decoder_outs[0]
			decoder_out = torch.log_softmax(decoder_out, dim=-1)
			return decoder_out, ys_pad_lens

			def inference(self,
			data_in,
			data_lengths=None,
			key: list = None,
			tokenizer=None,
			frontend=None,
			**kwargs,
			):

			def inference(
			self,
			data_in,
			data_lengths=None,
			key: list = None,
			tokenizer=None,
			frontend=None,
			**kwargs,
			):
			# init beamsearch

			is_use_ctc = kwargs.get("decoding_ctc_weight", 0.0) > 0.00001 and self.ctc != None
			is_use_lm = kwargs.get("lm_weight", 0.0) > 0.00001 and kwargs.get("lm_file", None) is not None
			is_use_lm = (
			kwargs.get("lm_weight", 0.0) > 0.00001 and kwargs.get("lm_file", None) is not None
			)
			if self.beam_search is None and (is_use_lm or is_use_ctc):
			logging.info("enable beam_search")
			self.init_beam_search(**kwargs)
			self.nbest = kwargs.get("nbest", 1)


			meta_data = {}


			# extract fbank feats
			time1 = time.perf_counter()
			audio_sample_list = load_audio_text_image_video(data_in, fs=frontend.fs, audio_fs=kwargs.get("fs", 16000))

			audio_sample_list = load_audio_text_image_video(
			data_in, fs=frontend.fs, audio_fs=kwargs.get("fs", 16000)
			)

			time2 = time.perf_counter()
			meta_data["load_data"] = f"{time2 - time1:0.3f}"
			speech, speech_lengths = extract_fbank(audio_sample_list, data_type=kwargs.get("data_type", "sound"),
			frontend=frontend)

			speech, speech_lengths = extract_fbank(
			audio_sample_list, data_type=kwargs.get("data_type", "sound"), frontend=frontend
			)
			time3 = time.perf_counter()
			meta_data["extract_feat"] = f"{time3 - time2:0.3f}"
			meta_data[
			"batch_data_time"] = speech_lengths.sum().item() * frontend.frame_shift * frontend.lfr_n / 1000

			meta_data["batch_data_time"] = (
			speech_lengths.sum().item() * frontend.frame_shift * frontend.lfr_n / 1000
			)

			speech = speech.to(device=kwargs["device"])
			speech_lengths = speech_lengths.to(device=kwargs["device"])

			# hotword
			self.hotword_list = self.generate_hotwords_list(kwargs.get("hotword", None), tokenizer=tokenizer, frontend=frontend)

			self.hotword_list = self.generate_hotwords_list(
			kwargs.get("hotword", None), tokenizer=tokenizer, frontend=frontend
			)

			# Encoder
			encoder_out, encoder_out_lens = self.encode(speech, speech_lengths)
			if isinstance(encoder_out, tuple):
			encoder_out = encoder_out[0]


			# predictor
			predictor_outs = self.calc_predictor(encoder_out, encoder_out_lens)
			pre_acoustic_embeds, pre_token_length, alphas, pre_peak_index = predictor_outs[0], predictor_outs[1], \
			predictor_outs[2], predictor_outs[3]
			pre_acoustic_embeds, pre_token_length, alphas, pre_peak_index = (
			predictor_outs[0],
			predictor_outs[1],
			predictor_outs[2],
			predictor_outs[3],
			)
			pre_token_length = pre_token_length.round().long()
			if torch.max(pre_token_length) < 1:
			return []


			decoder_outs = self.cal_decoder_with_predictor(encoder_out, encoder_out_lens,
			pre_acoustic_embeds,
			pre_token_length,
			hw_list=self.hotword_list,
			clas_scale=kwargs.get("clas_scale", 1.0))
			decoder_outs = self.cal_decoder_with_predictor(
			encoder_out,
			encoder_out_lens,
			pre_acoustic_embeds,
			pre_token_length,
			hw_list=self.hotword_list,
			clas_scale=kwargs.get("clas_scale", 1.0),
			)
			decoder_out, ys_pad_lens = decoder_outs[0], decoder_outs[1]


			results = []
			b, n, d = decoder_out.size()
			for i in range(b):
			x = encoder_out[i, :encoder_out_lens[i], :]
			am_scores = decoder_out[i, :pre_token_length[i], :]
			x = encoder_out[i, : encoder_out_lens[i], :]
			am_scores = decoder_out[i, : pre_token_length[i], :]
			if self.beam_search is not None:
			nbest_hyps = self.beam_search(
			x=x, am_scores=am_scores, maxlenratio=kwargs.get("maxlenratio", 0.0),
			minlenratio=kwargs.get("minlenratio", 0.0)
			x=x,
			am_scores=am_scores,
			maxlenratio=kwargs.get("maxlenratio", 0.0),
			minlenratio=kwargs.get("minlenratio", 0.0),
			)


			nbest_hyps = nbest_hyps[: self.nbest]
			else:


			yseq = am_scores.argmax(dim=-1)
			score = am_scores.max(dim=-1)[0]
			score = torch.sum(score, dim=-1)
			# pad with mask tokens to ensure compatibility with sos/eos tokens
			yseq = torch.tensor(
			[self.sos] + yseq.tolist() + [self.eos], device=yseq.device
			)
			yseq = torch.tensor([self.sos] + yseq.tolist() + [self.eos], device=yseq.device)
			nbest_hyps = [Hypothesis(yseq=yseq, score=score)]
			for nbest_idx, hyp in enumerate(nbest_hyps):
			ibest_writer = None
			@@ -381,26 +445,29 @@
			if not hasattr(self, "writer"):
			self.writer = DatadirWriter(kwargs.get("output_dir"))
			ibest_writer = self.writer[f"{nbest_idx + 1}best_recog"]


			# remove sos/eos and get results
			last_pos = -1
			if isinstance(hyp.yseq, list):
			token_int = hyp.yseq[1:last_pos]
			else:
			token_int = hyp.yseq[1:last_pos].tolist()


			# remove blank symbol id, which is assumed to be 0
			token_int = list(
			filter(lambda x: x != self.eos and x != self.sos and x != self.blank_id, token_int))

			filter(
			lambda x: x != self.eos and x != self.sos and x != self.blank_id, token_int
			)
			)

			if tokenizer is not None:
			# Change integer-ids to tokens
			token = tokenizer.ids2tokens(token_int)
			text = tokenizer.tokens2text(token)


			text_postprocessed, _ = postprocess_utils.sentence_postprocess(token)
			result_i = {"key": key[i], "text": text_postprocessed}


			if ibest_writer is not None:
			ibest_writer["token"][key[i]] = " ".join(token)
			ibest_writer["text"][key[i]] = text
			@@ -408,9 +475,8 @@
			else:
			result_i = {"key": key[i], "token_int": token_int}
			results.append(result_i)

			return results, meta_data

			return results, meta_data

			def generate_hotwords_list(self, hotword_list_or_file, tokenizer=None, frontend=None):
			def load_seg_dict(seg_dict_file):
			@@ -424,9 +490,9 @@
			value = s[1:]
			seg_dict[key] = " ".join(value)
			return seg_dict


			def seg_tokenize(txt, seg_dict):
			pattern = re.compile(r'^[\u4E00-\u9FA50-9]+$')
			pattern = re.compile(r"^[\u4E00-\u9FA50-9]+$")
			out_txt = ""
			for word in txt:
			word = word.lower()
			@@ -442,11 +508,11 @@
			else:
			out_txt += "<unk>" + " "
			return out_txt.strip().split()


			seg_dict = None
			if frontend.cmvn_file is not None:
			model_dir = os.path.dirname(frontend.cmvn_file)
			seg_dict_file = os.path.join(model_dir, 'seg_dict')
			seg_dict_file = os.path.join(model_dir, "seg_dict")
			if os.path.exists(seg_dict_file):
			seg_dict = load_seg_dict(seg_dict_file)
			else:
			@@ -455,11 +521,11 @@
			if hotword_list_or_file is None:
			hotword_list = None
			# for local txt inputs
			elif os.path.exists(hotword_list_or_file) and hotword_list_or_file.endswith('.txt'):
			elif os.path.exists(hotword_list_or_file) and hotword_list_or_file.endswith(".txt"):
			logging.info("Attempting to parse hotwords from local txt...")
			hotword_list = []
			hotword_str_list = []
			with codecs.open(hotword_list_or_file, 'r') as fin:
			with codecs.open(hotword_list_or_file, "r") as fin:
			for line in fin.readlines():
			hw = line.strip()
			hw_list = hw.split()
			@@ -468,11 +534,14 @@
			hotword_str_list.append(hw)
			hotword_list.append(tokenizer.tokens2ids(hw_list))
			hotword_list.append([self.sos])
			hotword_str_list.append('<s>')
			logging.info("Initialized hotword list from file: {}, hotword list: {}."
			.format(hotword_list_or_file, hotword_str_list))
			hotword_str_list.append("<s>")
			logging.info(
			"Initialized hotword list from file: {}, hotword list: {}.".format(
			hotword_list_or_file, hotword_str_list
			)
			)
			# for url, download and generate txt
			elif hotword_list_or_file.startswith('http'):
			elif hotword_list_or_file.startswith("http"):
			logging.info("Attempting to parse hotwords from url...")
			work_dir = tempfile.TemporaryDirectory().name
			if not os.path.exists(work_dir):
			@@ -483,7 +552,7 @@
			hotword_list_or_file = text_file_path
			hotword_list = []
			hotword_str_list = []
			with codecs.open(hotword_list_or_file, 'r') as fin:
			with codecs.open(hotword_list_or_file, "r") as fin:
			for line in fin.readlines():
			hw = line.strip()
			hw_list = hw.split()
			@@ -492,11 +561,14 @@
			hotword_str_list.append(hw)
			hotword_list.append(tokenizer.tokens2ids(hw_list))
			hotword_list.append([self.sos])
			hotword_str_list.append('<s>')
			logging.info("Initialized hotword list from file: {}, hotword list: {}."
			.format(hotword_list_or_file, hotword_str_list))
			hotword_str_list.append("<s>")
			logging.info(
			"Initialized hotword list from file: {}, hotword list: {}.".format(
			hotword_list_or_file, hotword_str_list
			)
			)
			# for text str input
			elif not hotword_list_or_file.endswith('.txt'):
			elif not hotword_list_or_file.endswith(".txt"):
			logging.info("Attempting to parse hotwords as str...")
			hotword_list = []
			hotword_str_list = []
			@@ -507,9 +579,19 @@
			hw_list = seg_tokenize(hw_list, seg_dict)
			hotword_list.append(tokenizer.tokens2ids(hw_list))
			hotword_list.append([self.sos])
			hotword_str_list.append('<s>')
			hotword_str_list.append("<s>")
			logging.info("Hotword list: {}.".format(hotword_str_list))
			else:
			hotword_list = None
			return hotword_list

			def export(
			self,
			**kwargs,
			):
			if "max_seq_len" not in kwargs:
			kwargs["max_seq_len"] = 512
			from .export_meta import export_rebuild_model

			models = export_rebuild_model(model=self, **kwargs)
			return models