update 8k uniasr recipe

仁迷

2023-02-07 6cc96a10eb7430e36e9b9b84a06fc8f29144e54e

update 8k uniasr recipe

 egs_modelscope/asr/uniasr/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-offline/finetune.py

@@ -25,7 +25,7 @@


if __name__ == '__main__':
    params = modelscope_args(model="damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online", data_path="./data")
    params = modelscope_args(model="damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-offline", data_path="./data")
    params.output_dir = "./checkpoint"              # m模型保存路径
    params.data_path = "./example_data/"            # 数据路径
    params.dataset_type = "small"                   # 小数据量设置small，若数据量大于1000小时，请使用large

 egs_modelscope/asr/uniasr/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-offline/infer.py

@@ -18,7 +18,7 @@
        os.environ['CUDA_VISIBLE_DEVICES'] = str(gpu_id)
    inference_pipline = pipeline(
        task=Tasks.auto_speech_recognition,
        model="damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online",
        model="damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-offline",
        output_dir=output_dir_job,
        batch_size=1
    )

 egs_modelscope/asr/uniasr/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-offline/infer_after_finetune.py

@@ -45,7 +45,7 @@

if __name__ == '__main__':
    params = {}
    params["modelscope_model_name"] = "damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online"
    params["modelscope_model_name"] = "damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-offline"
    params["required_files"] = ["am.mvn", "decoding.yaml", "configuration.json"]
    params["output_dir"] = "./checkpoint"
    params["data_dir"] = "./data/test"

 egs_modelscope/asr/uniasr/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online/finetune.py

@@ -25,7 +25,7 @@


if __name__ == '__main__':
    params = modelscope_args(model="damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-offline", data_path="./data")
    params = modelscope_args(model="damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online", data_path="./data")
    params.output_dir = "./checkpoint"              # m模型保存路径
    params.data_path = "./example_data/"            # 数据路径
    params.dataset_type = "small"                   # 小数据量设置small，若数据量大于1000小时，请使用large

 egs_modelscope/asr/uniasr/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online/infer.py

@@ -18,7 +18,7 @@
        os.environ['CUDA_VISIBLE_DEVICES'] = str(gpu_id)
    inference_pipline = pipeline(
        task=Tasks.auto_speech_recognition,
        model="damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-offline",
        model="damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online",
        output_dir=output_dir_job,
        batch_size=1
    )

 egs_modelscope/asr/uniasr/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online/infer_after_finetune.py

@@ -45,7 +45,7 @@

if __name__ == '__main__':
    params = {}
    params["modelscope_model_name"] = "damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-offline"
    params["modelscope_model_name"] = "damo/speech_UniASR_asr_2pass-zh-cn-8k-common-vocab3445-pytorch-online"
    params["required_files"] = ["am.mvn", "decoding.yaml", "configuration.json"]
    params["output_dir"] = "./checkpoint"
    params["data_dir"] = "./data/test"