From 580b11b57ac4b62f7e2acda73813a4e10e8e4cd3 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期二, 10 十月 2023 17:17:29 +0800
Subject: [PATCH] v0.8.0

---
 egs_modelscope/asr/TEMPLATE/README.md |   13 ++++++++-----
 1 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/egs_modelscope/asr/TEMPLATE/README.md b/egs_modelscope/asr/TEMPLATE/README.md
index a8cb486..bd0e6a9 100644
--- a/egs_modelscope/asr/TEMPLATE/README.md
+++ b/egs_modelscope/asr/TEMPLATE/README.md
@@ -27,15 +27,18 @@
 inference_pipeline = pipeline(
     task=Tasks.auto_speech_recognition,
     model='damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online',
-    model_revision='v1.0.6',
+    model_revision='v1.0.7',
     update_model=False,
     mode='paraformer_streaming'
     )
 import soundfile
 speech, sample_rate = soundfile.read("example/asr_example.wav")
 
-chunk_size = [5, 10, 5] #[5, 10, 5] 600ms, [8, 8, 4] 480ms
-param_dict = {"cache": dict(), "is_final": False, "chunk_size": chunk_size}
+chunk_size = [0, 10, 5] #[0, 10, 5] 600ms, [0, 8, 4] 480ms
+encoder_chunk_look_back = 4 #number of chunks to lookback for encoder self-attention
+decoder_chunk_look_back = 1 #number of encoder chunks to lookback for decoder cross-attention
+param_dict = {"cache": dict(), "is_final": False, "chunk_size": chunk_size,
+              "encoder_chunk_look_back": encoder_chunk_look_back, "decoder_chunk_look_back": decoder_chunk_look_back}
 chunk_stride = chunk_size[1] * 960 # 600ms銆�480ms
 # first chunk, 600ms
 speech_chunk = speech[0:chunk_stride]
@@ -55,7 +58,7 @@
 inference_pipeline = pipeline(
     task=Tasks.auto_speech_recognition,
     model='damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online',
-    model_revision='v1.0.6',
+    model_revision='v1.0.7',
     update_model=False,
     mode="paraformer_fake_streaming"
 )
@@ -199,7 +202,7 @@
 if __name__ == '__main__':
     params = modelscope_args(model="damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch", data_path="./data")
     params.output_dir = "./checkpoint"              # m妯″瀷淇濆瓨璺緞
-    params.data_path = "./example_data/"            # 鏁版嵁璺緞
+    params.data_path = "speech_asr_aishell1_trainsets"            # 鏁版嵁璺緞
     params.dataset_type = "small"                   # 灏忔暟鎹噺璁剧疆small锛岃嫢鏁版嵁閲忓ぇ浜�1000灏忔椂锛岃浣跨敤large
     params.batch_bins = 2000                       # batch size锛屽鏋渄ataset_type="small"锛宐atch_bins鍗曚綅涓篺bank鐗瑰緛甯ф暟锛屽鏋渄ataset_type="large"锛宐atch_bins鍗曚綅涓烘绉掞紝
     params.max_epoch = 20                           # 鏈�澶ц缁冭疆鏁�

--
Gitblit v1.9.1