From 0ee33ed14b24bf13ea70197c4b0cc6770e0d1639 Mon Sep 17 00:00:00 2001
From: 彭震东 <zhendong.peng@qq.com>
Date: 星期一, 18 三月 2024 21:56:20 +0800
Subject: [PATCH] fix typo (#1510)
---
funasr/models/paraformer_streaming/model.py | 23 ++++++++++-------------
1 files changed, 10 insertions(+), 13 deletions(-)
diff --git a/funasr/models/paraformer_streaming/model.py b/funasr/models/paraformer_streaming/model.py
index cebbfc1..a53aad5 100644
--- a/funasr/models/paraformer_streaming/model.py
+++ b/funasr/models/paraformer_streaming/model.py
@@ -531,6 +531,8 @@
for i in range(n):
kwargs["is_final"] = _is_final and i == n -1
audio_sample_i = audio_sample[i*chunk_stride_samples:(i+1)*chunk_stride_samples]
+ if kwargs["is_final"] and len(audio_sample_i) < 960:
+ continue
# extract fbank feats
speech, speech_lengths = extract_fbank([audio_sample_i], data_type=kwargs.get("data_type", "sound"),
@@ -566,7 +568,7 @@
max_seq_len=512,
**kwargs,
):
-
+ self.device = kwargs.get("device")
is_onnx = kwargs.get("type", "onnx") == "onnx"
encoder_class = tables.encoder_classes.get(kwargs["encoder"] + "Export")
self.encoder = encoder_class(self.encoder, onnx=is_onnx)
@@ -579,15 +581,10 @@
decoder_class = tables.decoder_classes.get(kwargs["decoder"] + "Export")
self.decoder = decoder_class(self.decoder, onnx=is_onnx)
- from funasr.utils.torch_function import MakePadMask
from funasr.utils.torch_function import sequence_mask
-
- if is_onnx:
- self.make_pad_mask = MakePadMask(max_seq_len, flip=False)
- else:
- self.make_pad_mask = sequence_mask(max_seq_len, flip=False)
-
- self.forward = self._export_forward
+
+
+ self.make_pad_mask = sequence_mask(max_seq_len, flip=False)
import copy
import types
@@ -595,7 +592,7 @@
decoder_model = copy.copy(self)
# encoder
- encoder_model.forward = types.MethodType(ParaformerStreaming._export_encoder_forward, encoder_model)
+ encoder_model.forward = types.MethodType(ParaformerStreaming.export_encoder_forward, encoder_model)
encoder_model.export_dummy_inputs = types.MethodType(ParaformerStreaming.export_encoder_dummy_inputs, encoder_model)
encoder_model.export_input_names = types.MethodType(ParaformerStreaming.export_encoder_input_names, encoder_model)
encoder_model.export_output_names = types.MethodType(ParaformerStreaming.export_encoder_output_names, encoder_model)
@@ -603,7 +600,7 @@
encoder_model.export_name = types.MethodType(ParaformerStreaming.export_encoder_name, encoder_model)
# decoder
- decoder_model.forward = types.MethodType(ParaformerStreaming._export_decoder_forward, decoder_model)
+ decoder_model.forward = types.MethodType(ParaformerStreaming.export_decoder_forward, decoder_model)
decoder_model.export_dummy_inputs = types.MethodType(ParaformerStreaming.export_decoder_dummy_inputs, decoder_model)
decoder_model.export_input_names = types.MethodType(ParaformerStreaming.export_decoder_input_names, decoder_model)
decoder_model.export_output_names = types.MethodType(ParaformerStreaming.export_decoder_output_names, decoder_model)
@@ -612,7 +609,7 @@
return encoder_model, decoder_model
- def _export_encoder_forward(
+ def export_encoder_forward(
self,
speech: torch.Tensor,
speech_lengths: torch.Tensor,
@@ -663,7 +660,7 @@
def export_encoder_name(self):
return "model.onnx"
- def _export_decoder_forward(
+ def export_decoder_forward(
self,
enc: torch.Tensor,
enc_len: torch.Tensor,
--
Gitblit v1.9.1