From e899096ce46ab74be7bdce64e24b91e86bb3be78 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期三, 11 十月 2023 16:19:52 +0800
Subject: [PATCH] Merge branch 'main' of github.com:alibaba-damo-academy/FunASR add

---
 egs_modelscope/asr/TEMPLATE/README.md |   40 +++++++++++++++++++++++++++++++---------
 1 files changed, 31 insertions(+), 9 deletions(-)

diff --git a/egs_modelscope/asr/TEMPLATE/README.md b/egs_modelscope/asr/TEMPLATE/README.md
index cf0ba84..e44a09d 100644
--- a/egs_modelscope/asr/TEMPLATE/README.md
+++ b/egs_modelscope/asr/TEMPLATE/README.md
@@ -1,3 +1,5 @@
+([绠�浣撲腑鏂嘳(./README_zh.md)|English)
+
 # Speech Recognition
 
 > **Note**:
@@ -25,15 +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]
@@ -53,7 +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"
 )
@@ -62,6 +67,23 @@
 print(rec_result)
 ```
 Full code of demo, please ref to [demo](https://github.com/alibaba-damo-academy/FunASR/discussions/241)
+
+#### [Paraformer-contextual Model](https://www.modelscope.cn/models/damo/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404/summary)
+```python
+from modelscope.pipelines import pipeline
+from modelscope.utils.constant import Tasks
+
+param_dict = dict()
+# param_dict['hotword'] = "https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/hotword.txt"
+param_dict['hotword']="閭撻儊鏉� 鐜嬮鏄� 鐜嬫檾鍚�"
+inference_pipeline = pipeline(
+    task=Tasks.auto_speech_recognition,
+    model="damo/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404",
+    param_dict=param_dict)
+
+rec_result = inference_pipeline(audio_in='https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_hotword.wav')
+print(rec_result)
+```
 
 #### [UniASR Model](https://www.modelscope.cn/models/damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online/summary)
 There are three decoding mode for UniASR model(`fast`銆乣normal`銆乣offline`), for more model details, please refer to [docs](https://www.modelscope.cn/models/damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online/summary)
@@ -197,7 +219,7 @@
 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�
@@ -230,10 +252,10 @@
     - `batch_bins`: batch size. For dataset_type is `small`, `batch_bins` indicates the feature frames. For dataset_type is `large`, `batch_bins` indicates the duration in ms
     - `max_epoch`: number of training epoch
     - `lr`: learning rate
-    - `init_param`: init model path, load modelscope model initialization by default. For example: ["checkpoint/20epoch.pb"]
-    - `freeze_param`: Freeze model parameters. For example锛歔"encoder"]
-    - `ignore_init_mismatch`: Ignore size mismatch when loading pre-trained model
-    - `use_lora`: Fine-tuning model use lora, more detail please refer to [LORA](https://arxiv.org/pdf/2106.09685.pdf)
+    - `init_param`: `[]`(Default), init model path, load modelscope model initialization by default. For example: ["checkpoint/20epoch.pb"]
+    - `freeze_param`: `[]`(Default), Freeze model parameters. For example锛歔"encoder"]
+    - `ignore_init_mismatch`: `True`(Default), Ignore size mismatch when loading pre-trained model
+    - `use_lora`: `False`(Default), Fine-tuning model use lora, more detail please refer to [LORA](https://arxiv.org/pdf/2106.09685.pdf)
 
 - Training data formats锛�
 ```sh

--
Gitblit v1.9.1