ValueError：调用层“transformer”时遇到异常（变压器类型）

Question

因此，我编写了一个作为 ASR 工作的 Transformers 神经网络，它工作正常，训练良好并将模型保存为...

model.save("savedmodel.model")

问题是，当我想要预测时，我这样做...

speech_model = load_model('D:\DOT\Speechrecognition\speechrecognitionE.model')

path = "D:\DOT\Speechrecognition\Data\LJSpeech-1.1\wavs\LJ001-0001.wav"

def path_to_audio(path):
    # spectrogram using stft
    audio = tf.io.read_file(path)
    audio, _ = tf.audio.decode_wav(audio, 1)
    audio = tf.squeeze(audio, axis=-1)
    stfts = tf.signal.stft(audio, frame_length=200, frame_step=80, fft_length=256)
    x = tf.math.pow(tf.abs(stfts), 0.5)
    # normalisation
    means = tf.math.reduce_mean(x, 1, keepdims=True)
    stddevs = tf.math.reduce_std(x, 1, keepdims=True)
    x = (x - means) / stddevs
    audio_len = tf.shape(x)[0]
    # padding to 10 seconds
    pad_len = 2754
    paddings = tf.constant([[0, pad_len], [0, 0]])
    x = tf.pad(x, paddings, "CONSTANT")[:pad_len, :]
    return x

x = path_to_audio(path)
#print(x)
speech_model.predict(x)

音频函数的路径，将频谱图的音频路径，在训练模型中它接收音频频谱图作为输入，但它显示此错误。

    Traceback (most recent call last):
File "C:\Users\berna\Desktop\Programming\AI_ML_DL\Projects\DOT\DOT-alpha.py", line 72, in <module>
    speech_model.predict(x)
File "C:\Users\berna\AppData\Roaming\Python\Python39\site-packages\keras\utils\traceback_utils.py", line 67, in error_handler
    raise e.with_traceback(filtered_tb) from None
File "C:\Users\berna\AppData\Roaming\Python\Python39\site-packages\tensorflow\python\framework\func_graph.py", line 1129, in autograph_handler
    raise e.ag_error_metadata.to_exception(e)
ValueError: in user code:

    File "C:\Users\berna\AppData\Roaming\Python\Python39\site-packages\keras\engine\training.py", line 1621, in predict_function  *
        return step_function(self, iterator)
    File "C:\Users\berna\AppData\Roaming\Python\Python39\site-packages\keras\engine\training.py", line 1611, in step_function  **
        outputs = model.distribute_strategy.run(run_step, args=(data,))
    File "C:\Users\berna\AppData\Roaming\Python\Python39\site-packages\keras\engine\training.py", line 1604, in run_step  **
        outputs = model.predict_step(data)
    File "C:\Users\berna\AppData\Roaming\Python\Python39\site-packages\keras\engine\training.py", line 1572, in predict_step
        return self(x, training=False)
    File "C:\Users\berna\AppData\Roaming\Python\Python39\site-packages\keras\utils\traceback_utils.py", line 67, in error_handler
        raise e.with_traceback(filtered_tb) from None

    ValueError: Exception encountered when calling layer "transformer" (type Transformer).

    Could not find matching concrete function to call loaded from the SavedModel. Got:
    Positional arguments (2 total):
        * Tensor("inputs:0", shape=(None, 129), dtype=float32)
        * False
    Keyword arguments: {}

    Expected these arguments to match one of the following 4 option(s):

    Option 1:
    Positional arguments (2 total):
        * [TensorSpec(shape=(None, None, 129), dtype=tf.float32, name='inputs/0'), TensorSpec(shape=(None, 199), dtype=tf.int32, name='inputs/1')]
        * False
    Keyword arguments: {}

    Option 2:
    Positional arguments (2 total):
        * [TensorSpec(shape=(None, None, 129), dtype=tf.float32, name='inputs/0'), TensorSpec(shape=(None, 199), dtype=tf.int32, name='inputs/1')]
        * True
    Keyword arguments: {}

    Option 3:
    Positional arguments (2 total):
        * [TensorSpec(shape=(None, None, 129), dtype=tf.float32, name='input_1'), TensorSpec(shape=(None, 199), dtype=tf.int32, name='input_2')]
        * False
    Keyword arguments: {}

    Option 4:
    Positional arguments (2 total):
        * [TensorSpec(shape=(None, None, 129), dtype=tf.float32, name='input_1'), TensorSpec(shape=(None, 199), dtype=tf.int32, name='input_2')]
        * True
    Keyword arguments: {}

    Call arguments received:
    • args=('tf.Tensor(shape=(None, 129), dtype=float32)',)
    • kwargs={'training': 'False'}

这意味着什么？预测有什么问题？