Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
update _hw_repre
shixian.shi66d3b5c21277
diff --git a/funasr/models/seaco_paraformer/model.py b/funasr/models/seaco_paraformer/model.pyindex caf2b15c7..0287f56ed 100644--- a/funasr/models/seaco_paraformer/model.py+++ b/funasr/models/seaco_paraformer/model.py@@ -19,11 +19,9 @@ from distutils.version import LooseVersionfrom funasr.register import tablesfrom funasr.utils import postprocess_utils-from funasr.metrics.compute_acc import th_accuracyfrom funasr.models.paraformer.model import Paraformerfrom funasr.utils.datadir_writer import DatadirWriterfrom funasr.models.paraformer.search import Hypothesis-from funasr.models.paraformer.cif_predictor import mae_lossfrom funasr.train_utils.device_funcs import force_gatherablefrom funasr.models.bicif_paraformer.model import BiCifParaformerfrom funasr.losses.label_smoothing_loss import LabelSmoothingLoss@@ -76,7 +74,7 @@ class SeacoParaformer(BiCifParaformer, Paraformer):self.lstm_proj = torch.nn.Linear(self.inner_dim*2, self.inner_dim)else:self.lstm_proj = None- self.bias_embed = torch.nn.Embedding(self.vocab_size, self.inner_dim)+ # self.bias_embed = torch.nn.Embedding(self.vocab_size, self.inner_dim)elif self.bias_encoder_type == 'mean':self.bias_embed = torch.nn.Embedding(self.vocab_size, self.inner_dim)else:@@ -274,6 +272,8 @@ class SeacoParaformer(BiCifParaformer, Paraformer):hotword_lengths):if self.bias_encoder_type != 'lstm':logging.error("Unsupported bias encoder type")++ '''hw_embed = self.decoder.embed(hotword_pad)hw_embed, (_, _) = self.bias_encoder(hw_embed)if self.lstm_proj is not None:@@ -281,26 +281,20 @@ class SeacoParaformer(BiCifParaformer, Paraformer):_ind = np.arange(0, hw_embed.shape[0]).tolist()selected = hw_embed[_ind, [i-1 for i in hotword_lengths.detach().cpu().tolist()]]return selected+ '''- '''- def calc_predictor(self, encoder_out, encoder_out_lens):- encoder_out_mask = (~make_pad_mask(encoder_out_lens, maxlen=encoder_out.size(1))[:, None, :]).to(- encoder_out.device)- pre_acoustic_embeds, pre_token_length, alphas, pre_peak_index, pre_token_length2 = self.predictor(encoder_out,- None,- encoder_out_mask,- ignore_id=self.ignore_id)- return pre_acoustic_embeds, pre_token_length, alphas, pre_peak_index--- def calc_predictor_timestamp(self, encoder_out, encoder_out_lens, token_num):- encoder_out_mask = (~make_pad_mask(encoder_out_lens, maxlen=encoder_out.size(1))[:, None, :]).to(- encoder_out.device)- ds_alphas, ds_cif_peak, us_alphas, us_peaks = self.predictor.get_upsample_timestamp(encoder_out,- encoder_out_mask,- token_num)- return ds_alphas, ds_cif_peak, us_alphas, us_peaks- '''+ # hw_embed = self.sac_embedding(hotword_pad)+ hw_embed = self.decoder.embed(hotword_pad)+ hw_embed = torch.nn.utils.rnn.pack_padded_sequence(hw_embed, hotword_lengths.cpu().type(torch.int64), batch_first=True, enforce_sorted=False)+ packed_rnn_output, _ = self.bias_encoder(hw_embed)+ rnn_output = torch.nn.utils.rnn.pad_packed_sequence(packed_rnn_output, batch_first=True)[0]+ if self.lstm_proj is not None:+ hw_hidden = self.lstm_proj(rnn_output)+ else:+ hw_hidden = rnn_output+ _ind = np.arange(0, hw_hidden.shape[0]).tolist()+ selected = hw_hidden[_ind, [i-1 for i in hotword_lengths.detach().cpu().tolist()]]+ return selecteddef inference(self,data_in,