install requirements automatically

2026-10-12 02:19:18 +00:00 · 2024-03-25 12:37:35 +08:00
parent 8c1016ca77
commit 447222c00e
50 changed files with 72 additions and 72 deletions
@@ -12,7 +12,7 @@ multilingual_wavs = [
    "example_ko.mp3",
 ]

-model = AutoModel(model="iic/speech_whisper-large_lid_multilingual_pytorch", model_revision="v2.0.4")
+model = AutoModel(model="iic/speech_whisper-large_lid_multilingual_pytorch", model_revision="master")
 for wav_id in multilingual_wavs:
    wav_file = f"{model.model_path}/examples/{wav_id}"
    res = model.generate(input=wav_file, data_type="sound", inference_clip_length=250)
@@ -15,7 +15,7 @@ multilingual_wavs=[

 inference_pipeline = pipeline(
    task=Tasks.auto_speech_recognition,
-    model='iic/speech_whisper-large_lid_multilingual_pytorch', model_revision="v2.0.4")
+    model='iic/speech_whisper-large_lid_multilingual_pytorch', model_revision="master")

 for wav in multilingual_wavs:
    rec_result = inference_pipeline(input=wav, inference_clip_length=250)
@@ -6,11 +6,11 @@
 from funasr import AutoModel

 model = AutoModel(model="iic/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch",
-                  model_revision="v2.0.4",
+                  model_revision="master",
                  vad_model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch",
-                  vad_model_revision="v2.0.4",
+                  vad_model_revision="master",
                  punc_model="iic/punc_ct-transformer_cn-en-common-vocab471067-large",
-                  punc_model_revision="v2.0.4",
+                  punc_model_revision="master",
                  # spk_model="iic/speech_campplus_sv_zh-cn_16k-common",
                  # spk_model_revision="v2.0.2",
                  )
@@ -1,11 +1,11 @@

 model="iic/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
-model_revision="v2.0.4"
+model_revision="master"
 vad_model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch"
-vad_model_revision="v2.0.4"
+vad_model_revision="master"
 #punc_model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
 punc_model="iic/punc_ct-transformer_cn-en-common-vocab471067-large"
-punc_model_revision="v2.0.4"
+punc_model_revision="master"
 spk_model="iic/speech_campplus_sv_zh-cn_16k-common"
 spk_model_revision="v2.0.2"

@@ -8,7 +8,7 @@
 from funasr import AutoModel

 model = AutoModel(model="iic/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch",
-                  model_revision="v2.0.4", device="cpu")
+                  model_revision="master", device="cpu")

 res = model.export(type="onnx", quantize=False)
 print(res)
@@ -6,7 +6,7 @@ export HYDRA_FULL_ERROR=1


 model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
-model_revision="v2.0.4"
+model_revision="master"

 python -m funasr.bin.export \
 ++model=${model} \
@@ -11,7 +11,7 @@ gpu_num=$(echo $CUDA_VISIBLE_DEVICES | awk -F "," '{print NF}')

 ## option 1, download model automatically
 model_name_or_model_dir="iic/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
-model_revision="v2.0.4"
+model_revision="master"

 ## option 2, download model by git
 #local_path_root=${workspace}/modelscope_models
@@ -5,7 +5,7 @@

 from funasr import AutoModel

-model = AutoModel(model="iic/speech_conformer_asr_nat-zh-cn-16k-aishell2-vocab5212-pytorch", model_revision="v2.0.4",
+model = AutoModel(model="iic/speech_conformer_asr_nat-zh-cn-16k-aishell2-vocab5212-pytorch", model_revision="master",
                  )

 res = model.generate(input="https://modelscope.oss-cn-beijing.aliyuncs.com/test/audios/asr_example.wav")
@@ -1,6 +1,6 @@

 model="iic/speech_conformer_asr_nat-zh-cn-16k-aishell1-vocab4234-pytorch"
-model_revision="v2.0.4"
+model_revision="master"

 python funasr/bin/inference.py \
 +model=${model} \
@@ -5,7 +5,7 @@

 from funasr import AutoModel

-model = AutoModel(model="iic/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404", model_revision="v2.0.4")
+model = AutoModel(model="iic/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404", model_revision="master")

 res = model.generate(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav",
            hotword='达摩院 魔搭')
@@ -1,6 +1,6 @@

 model="iic/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404"
-model_revision="v2.0.4"
+model_revision="master"

 python ../../../funasr/bin/inference.py \
 +model=${model} \
@@ -11,7 +11,7 @@ gpu_num=$(echo $CUDA_VISIBLE_DEVICES | awk -F "," '{print NF}')

 ## option 1, download model automatically
 model_name_or_model_dir="iic/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404"
-model_revision="v2.0.4"
+model_revision="master"

 ## option 2, download model by git
 #local_path_root=${workspace}/modelscope_models
@@ -5,7 +5,7 @@

 from funasr import AutoModel

-model = AutoModel(model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch", model_revision="v2.0.4")
+model = AutoModel(model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch", model_revision="master")

 res = model.generate(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_text/punc_example.txt")
 print(res)
@@ -13,7 +13,7 @@ print(res)

 from funasr import AutoModel

-model = AutoModel(model="iic/punc_ct-transformer_cn-en-common-vocab471067-large", model_revision="v2.0.4")
+model = AutoModel(model="iic/punc_ct-transformer_cn-en-common-vocab471067-large", model_revision="master")

 res = model.generate(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_text/punc_example.txt")
 print(res)
@@ -1,9 +1,9 @@

 #model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
-#model_revision="v2.0.4"
+#model_revision="master"

 model="iic/punc_ct-transformer_cn-en-common-vocab471067-large"
-model_revision="v2.0.4"
+model_revision="master"

 python funasr/bin/inference.py \
 +model=${model} \
@@ -8,7 +8,7 @@
 from funasr import AutoModel

 model = AutoModel(model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch",
-                  model_revision="v2.0.4")
+                  model_revision="master")

 res = model.export(type="onnx", quantize=False)
 print(res)
@@ -6,7 +6,7 @@ export HYDRA_FULL_ERROR=1


 model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
-model_revision="v2.0.4"
+model_revision="master"

 python -m funasr.bin.export \
 ++model=${model} \
@@ -5,7 +5,7 @@

 from funasr import AutoModel

-model = AutoModel(model="iic/punc_ct-transformer_zh-cn-common-vad_realtime-vocab272727", model_revision="v2.0.4")
+model = AutoModel(model="iic/punc_ct-transformer_zh-cn-common-vad_realtime-vocab272727", model_revision="master")

 inputs = "跨境河流是养育沿岸|人民的生命之源长期以来为帮助下游地区防灾减灾中方技术人员|在上游地区极为恶劣的自然条件下克服巨大困难甚至冒着生命危险|向印方提供汛期水文资料处理紧急事件中方重视印方在跨境河流问题上的关切|愿意进一步完善双方联合工作机制|凡是|中方能做的我们|都会去做而且会做得更好我请印度朋友们放心中国在上游的|任何开发利用都会经过科学|规划和论证兼顾上下游的利益"
 vads = inputs.split("|")
@@ -1,6 +1,6 @@

 model="iic/punc_ct-transformer_zh-cn-common-vad_realtime-vocab272727"
-model_revision="v2.0.4"
+model_revision="master"

 python funasr/bin/inference.py \
 +model=${model} \
@@ -8,7 +8,7 @@
 from funasr import AutoModel

 model = AutoModel(model="iic/punc_ct-transformer_zh-cn-common-vad_realtime-vocab272727",
-                  model_revision="v2.0.4")
+                  model_revision="master")

 res = model.export(type="onnx", quantize=False)
 print(res)
@@ -6,7 +6,7 @@ export HYDRA_FULL_ERROR=1


 model="iic/punc_ct-transformer_zh-cn-common-vad_realtime-vocab272727"
-model_revision="v2.0.4"
+model_revision="master"

 python -m funasr.bin.export \
 ++model=${model} \
@@ -6,9 +6,9 @@
 from funasr import AutoModel

 # model="iic/emotion2vec_base"
-model = AutoModel(model="iic/emotion2vec_base_finetuned", model_revision="v2.0.4",
+model = AutoModel(model="iic/emotion2vec_base_finetuned", model_revision="master",
                  # vad_model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch",
-                  # vad_model_revision="v2.0.4",
+                  # vad_model_revision="master",
                  # vad_kwargs={"max_single_segment_time": 2000},
                  )

@@ -6,7 +6,7 @@
 from funasr import AutoModel
 wav_file = "https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/vad_example.wav"

-model = AutoModel(model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch", model_revision="v2.0.4")
+model = AutoModel(model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch", model_revision="master")

 res = model.generate(input=wav_file)
 print(res)
@@ -1,7 +1,7 @@


 model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch"
-model_revision="v2.0.4"
+model_revision="master"

 python funasr/bin/inference.py \
 +model=${model} \
@@ -8,7 +8,7 @@

 from funasr import AutoModel

-model = AutoModel(model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch", model_revision="v2.0.4")
+model = AutoModel(model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch", model_revision="master")

 res = model.export(type="onnx", quantize=False)
 print(res)
@@ -7,7 +7,7 @@ export HYDRA_FULL_ERROR=1


 model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch"
-model_revision="v2.0.4"
+model_revision="master"

 python -m funasr.bin.export \
 ++model=${model} \
@@ -5,7 +5,7 @@

 from funasr import AutoModel

-model = AutoModel(model="iic/speech_timestamp_prediction-v1-16k-offline", model_revision="v2.0.4")
+model = AutoModel(model="iic/speech_timestamp_prediction-v1-16k-offline", model_revision="master")

 res = model.generate(input=("https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav",
                   "欢迎大家来到魔搭社区进行体验"),
@@ -1,6 +1,6 @@

 model="iic/speech_timestamp_prediction-v1-16k-offline"
-model_revision="v2.0.4"
+model_revision="master"

 python funasr/bin/inference.py \
 +model=${model} \
@@ -6,11 +6,11 @@
 from funasr import AutoModel

 model = AutoModel(model="iic/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch",
-                  model_revision="v2.0.4",
+                  model_revision="master",
                  vad_model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch",
-                  vad_model_revision="v2.0.4",
+                  vad_model_revision="master",
                  punc_model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch",
-                  punc_model_revision="v2.0.4",
+                  punc_model_revision="master",
                  # spk_model="iic/speech_campplus_sv_zh-cn_16k-common",
                  # spk_model_revision="v2.0.2"
                  )
@@ -1,10 +1,10 @@

 model="iic/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
-model_revision="v2.0.4"
+model_revision="master"
 vad_model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch"
-vad_model_revision="v2.0.4"
+vad_model_revision="master"
 punc_model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
-punc_model_revision="v2.0.4"
+punc_model_revision="master"
 spk_model="iic/speech_campplus_sv_zh-cn_16k-common"
 spk_model_revision="v2.0.2"

@@ -6,12 +6,12 @@
 from funasr import AutoModel

 model = AutoModel(model="iic/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch", 
-                  model_revision="v2.0.4",
+                  model_revision="master",
                  vad_model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch",
-                  vad_model_revision="v2.0.4",
+                  vad_model_revision="master",
                  vad_kwargs={"max_single_segment_time": 60000},
                  punc_model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch",
-                  punc_model_revision="v2.0.4",
+                  punc_model_revision="master",
                  # spk_model="iic/speech_campplus_sv_zh-cn_16k-common",
                  # spk_model_revision="v2.0.2",
                  )
@@ -23,7 +23,7 @@ print(res)
 ''' can not use currently
 from funasr import AutoFrontend

-frontend = AutoFrontend(model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch", model_revision="v2.0.4")
+frontend = AutoFrontend(model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch", model_revision="master")

 fbanks = frontend(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav", batch_size=2)

@@ -10,7 +10,7 @@
 from funasr import AutoModel

 model = AutoModel(model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch",
-                  model_revision="v2.0.4")
+                  model_revision="master")

 res = model.export(type="onnx", quantize=False)
 print(res)
@@ -6,7 +6,7 @@ export HYDRA_FULL_ERROR=1


 model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
-model_revision="v2.0.4"
+model_revision="master"


 python -m funasr.bin.export \
@@ -10,7 +10,7 @@ gpu_num=$(echo $CUDA_VISIBLE_DEVICES | awk -F "," '{print NF}')

 ## option 1, download model automatically
 model_name_or_model_dir="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
-model_revision="v2.0.4"
+model_revision="master"

 ## option 2, download model by git
 #local_path_root=${workspace}/modelscope_models
@@ -9,7 +9,7 @@ input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr
 output_dir="./outputs/debug"

 model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
-model_revision="v2.0.4"
+model_revision="master"

 device="cuda:0" # "cuda:0" for gpu0, "cuda:1" for gpu1, "cpu"

@@ -10,7 +10,7 @@ from funasr import AutoModel
 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
-model = AutoModel(model="iic/speech_paraformer_asr_nat-zh-cn-16k-common-vocab8404-online", model_revision="v2.0.4")
+model = AutoModel(model="iic/speech_paraformer_asr_nat-zh-cn-16k-common-vocab8404-online", model_revision="master")

 wav_file = os.path.join(model.model_path, "example/asr_example.wav")
 res = model.generate(input=wav_file,
@@ -1,6 +1,6 @@

 model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online"
-model_revision="v2.0.4"
+model_revision="master"

 python funasr/bin/inference.py \
 +model=${model} \
@@ -10,7 +10,7 @@
 from funasr import AutoModel

 model = AutoModel(model="iic/speech_paraformer_asr_nat-zh-cn-16k-common-vocab8404-online",
-                  model_revision="v2.0.4")
+                  model_revision="master")

 res = model.export(type="onnx", quantize=False)
 print(res)
@@ -6,7 +6,7 @@ export HYDRA_FULL_ERROR=1


 model="iic/speech_paraformer_asr_nat-zh-cn-16k-common-vocab8404-online"
-model_revision="v2.0.4"
+model_revision="master"


 python -m funasr.bin.export \
@@ -10,7 +10,7 @@ gpu_num=$(echo $CUDA_VISIBLE_DEVICES | awk -F "," '{print NF}')

 ## option 1, download model automatically
 model_name_or_model_dir="iic/speech_paraformer_asr_nat-zh-cn-16k-common-vocab8404-online"
-model_revision="v2.0.4"
+model_revision="master"

 ## option 2, download model by git
 #local_path_root=${workspace}/modelscope_models
@@ -9,7 +9,7 @@ chunk_size = [5, 10, 5] #[0, 10, 5] 600ms, [0, 8, 4] 480ms
 encoder_chunk_look_back = 0 #number of chunks to lookback for encoder self-attention
 decoder_chunk_look_back = 0 #number of encoder chunks to lookback for decoder cross-attention

-model = AutoModel(model="/Users/zhifu/Downloads/modelscope_models/speech_SCAMA_asr-zh-cn-16k-common-vocab8358-streaming", model_revision="v2.0.4")
+model = AutoModel(model="/Users/zhifu/Downloads/modelscope_models/speech_SCAMA_asr-zh-cn-16k-common-vocab8358-streaming", model_revision="master")
 cache = {}
 res = model.generate(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav",
            chunk_size=chunk_size,
@@ -1,6 +1,6 @@

 model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online"
-model_revision="v2.0.4"
+model_revision="master"

 python funasr/bin/inference.py \
 +model=${model} \
@@ -6,11 +6,11 @@
 from funasr import AutoModel

 model = AutoModel(model="iic/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch",
-                  model_revision="v2.0.4",
+                  model_revision="master",
                  # vad_model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch",
-                  # vad_model_revision="v2.0.4",
+                  # vad_model_revision="master",
                  # punc_model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch",
-                  # punc_model_revision="v2.0.4",
+                  # punc_model_revision="master",
                  # spk_model="iic/speech_campplus_sv_zh-cn_16k-common",
                  # spk_model_revision="v2.0.2",
                  )
@@ -1,10 +1,10 @@

 model="iic/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
-model_revision="v2.0.4"
+model_revision="master"
 vad_model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch"
-vad_model_revision="v2.0.4"
+vad_model_revision="master"
 punc_model="iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
-punc_model_revision="v2.0.4"
+punc_model_revision="master"

 python funasr/bin/inference.py \
 +model=${model} \
@@ -6,7 +6,7 @@
 from funasr import AutoModel


-model = AutoModel(model="iic/speech_UniASR-large_asr_2pass-zh-cn-16k-common-vocab8358-tensorflow1-offline", model_revision="v2.0.4",)
+model = AutoModel(model="iic/speech_UniASR-large_asr_2pass-zh-cn-16k-common-vocab8358-tensorflow1-offline", model_revision="master",)


 res = model.generate(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav")
@@ -16,7 +16,7 @@ print(res)
 ''' can not use currently
 from funasr import AutoFrontend

-frontend = AutoFrontend(model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch", model_revision="v2.0.4")
+frontend = AutoFrontend(model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch", model_revision="master")

 fbanks = frontend(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav", batch_size=2)

@@ -1,6 +1,6 @@

 model="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
-model_revision="v2.0.4"
+model_revision="master"

 python funasr/bin/inference.py \
 +model=${model} \
@@ -11,7 +11,7 @@ input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr
 output_dir="./outputs/debug"

 model="iic/speech_whisper-large_asr_multilingual"
-model_revision="v2.0.4"
+model_revision="master"

 device="cuda:0" # "cuda:0" for gpu0, "cuda:1" for gpu1, "cpu"

@@ -109,7 +109,7 @@ class AutoModel:
        if vad_model is not None:
            logging.info("Building VAD model.")
            vad_kwargs["model"] = vad_model
-            vad_kwargs["model_revision"] = kwargs.get("vad_model_revision", None)
+            vad_kwargs["model_revision"] = kwargs.get("vad_model_revision", "master")
            vad_kwargs["device"] = kwargs["device"]
            vad_model, vad_kwargs = self.build_model(**vad_kwargs)

@@ -119,7 +119,7 @@ class AutoModel:
        if punc_model is not None:
            logging.info("Building punc model.")
            punc_kwargs["model"] = punc_model
-            punc_kwargs["model_revision"] = kwargs.get("punc_model_revision", None)
+            punc_kwargs["model_revision"] = kwargs.get("punc_model_revision", "master")
            punc_kwargs["device"] = kwargs["device"]
            punc_model, punc_kwargs = self.build_model(**punc_kwargs)

@@ -129,7 +129,7 @@ class AutoModel:
        if spk_model is not None:
            logging.info("Building SPK model.")
            spk_kwargs["model"] = spk_model
-            spk_kwargs["model_revision"] = kwargs.get("spk_model_revision", None)
+            spk_kwargs["model_revision"] = kwargs.get("spk_model_revision", "master")
            spk_kwargs["device"] = kwargs["device"]
            spk_model, spk_kwargs = self.build_model(**spk_kwargs)
            self.cb_model = ClusterBackend().to(kwargs["device"])
@@ -29,7 +29,7 @@ def download_from_ms(**kwargs):
    model_or_path = kwargs.get("model")
    if model_or_path in name_maps_ms:
        model_or_path = name_maps_ms[model_or_path]
-    model_revision = kwargs.get("model_revision")
+    model_revision = kwargs.get("model_revision", "master")
    if not os.path.exists(model_or_path) and "model_path" not in kwargs:
        try:
            model_or_path = get_or_download_model_dir(model_or_path, model_revision,
@@ -73,7 +73,7 @@ class LLMASR(nn.Module):
        hub = audio_encoder_conf.get("hub", None)
        if hub == "ms":
            from funasr import AutoModel
-            model = AutoModel(model=audio_encoder, model_revision="v2.0.4")
+            model = AutoModel(model=audio_encoder, model_revision="master")
            # frontend = model.kwargs.get("frontend")
            audio_encoder_output_size = model.model.encoder_output_size

@@ -75,7 +75,7 @@ class LLMASRNAR(nn.Module):
        if hub == "funasr":
            from funasr import AutoModel
            init_param_path = encoder_conf.get("init_param_path", "iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch")
-            model = AutoModel(model=init_param_path, model_revision="v2.0.4")
+            model = AutoModel(model=init_param_path, model_revision="master")
            # frontend = model.kwargs.get("frontend")
            model.model.decoder = None
            
@@ -406,7 +406,7 @@ class LLMASRNARPrompt(nn.Module):
            from funasr import AutoModel
            init_param_path = encoder_conf.get("init_param_path",
                                               "iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch")
-            model = AutoModel(model=init_param_path, model_revision="v2.0.4")
+            model = AutoModel(model=init_param_path, model_revision="master")
            # frontend = model.kwargs.get("frontend")
            model.model.decoder = None