paddlespeech - list index out of range

[zhanghzong]

环境

1. win11
2. conda Python 3.10.11
3. pip 23.1.2
4. paddlepaddle==2.5.0
5. paddlespeech==1.4.1
6. paddleaudio==1.0.1
7. vs 2022

使用

参照源 https://github.com/PaddlePaddle/PaddleSpeech/blob/develop/demos/speech_recognition/README_cn.md

代码

import paddle
from paddlespeech.cli.asr import ASRExecutor

asr_executor = ASRExecutor()
text = asr_executor(
    model='conformer_wenetspeech',
    lang='zh',
    sample_rate=16000,
    config=None,  # Set `config` and `ckpt_path` to None to use pretrained model.
    ckpt_path=None,
    audio_file='./zh.wav',
    force_yes=False,
    device=paddle.get_device())
print('ASR Result: \n{}'.format(text))

错误详情

2023-07-21 11:09:44.658 | INFO     | paddlespeech.s2t.modules.ctc:<module>:45 - paddlespeech_ctcdecoders not installed!
2023-07-21 11:09:44.734 | INFO     | paddlespeech.s2t.modules.embedding:__init__:150 - max len: 5000
[2023-07-21 11:09:48,553] [   ERROR] - list index out of range
Traceback (most recent call last):
  File "F:\Program\anaconda3\envs\paddle.1\lib\site-packages\paddlespeech\cli\asr\infer.py", line 314, in infer
    result_transcripts = self.model.decode(
  File "F:\Program\anaconda3\envs\paddle.1\lib\site-packages\decorator.py", line 232, in fun
    return caller(func, *(extras + args), **kw)
  File "F:\Program\anaconda3\envs\paddle.1\lib\site-packages\paddle\fluid\dygraph\base.py", line 347, in _decorate_function
    return func(*args, **kwargs)
  File "F:\Program\anaconda3\envs\paddle.1\lib\site-packages\paddlespeech\s2t\models\u2\u2.py", line 818, in decode
    hyp = self.attention_rescoring(
  File "F:\Program\anaconda3\envs\paddle.1\lib\site-packages\paddlespeech\s2t\models\u2\u2.py", line 532, in attention_rescoring
    assert speech.shape[0] == speech_lengths.shape[0]
IndexError: list index out of range
Traceback (most recent call last):

报错源码

    def attention_rescoring(self,
                            speech: paddle.Tensor,
                            speech_lengths: paddle.Tensor,
                            beam_size: int,
                            decoding_chunk_size: int=-1,
                            num_decoding_left_chunks: int=-1,
                            ctc_weight: float=0.0,
                            simulate_streaming: bool=False,
                            reverse_weight: float=0.0) -> List[int]:

        assert speech.shape[0] == speech_lengths.shape[0]

[Chuyaoyuan]

尝试下直接编译paddlespeech的最新develop分支吧

[zhanghzong]

尝试下直接编译paddlespeech的最新develop分支吧

试试看

[zhanghzong]

特来补充说明. 编译 develop 分支 可以解决上述问题.

感谢 @Chuyaoyuan 提示

[zhanghzong]

新的问题

开发分支: develop 存在以下问题

Traceback (most recent call last):
  File "F:\Program\anaconda3\envs\paddle.1\Scripts\paddlespeech-script.py", line 33, in <module>
    sys.exit(load_entry_point('paddlespeech==0.0.0', 'console_scripts', 'paddlespeech')())
  File "F:\Program\anaconda3\envs\paddle.1\lib\site-packages\paddlespeech\cli\entry.py", line 40, in _execute
    exec("from {} import {}".format(module, cls))
  File "<string>", line 1, in <module>
  File "F:\Program\anaconda3\envs\paddle.1\lib\site-packages\paddlespeech\cli\tts\__init__.py", line 14, in <module>
    from .infer import TTSExecutor
  File "F:\Program\anaconda3\envs\paddle.1\lib\site-packages\paddlespeech\cli\tts\infer.py", line 33, in <module>
    from paddlespeech.t2s.exps.syn_utils import get_am_inference
  File "F:\Program\anaconda3\envs\paddle.1\lib\site-packages\paddlespeech\t2s\exps\syn_utils.py", line 36, in <module>
    from paddlespeech.t2s.frontend.canton_frontend import CantonFrontend
  File "F:\Program\anaconda3\envs\paddle.1\lib\site-packages\paddlespeech\t2s\frontend\canton_frontend.py", line 19, in <module>
    import ToJyutping
  File "f:\program\anaconda3\envs\paddle.1\lib\site-packages\ToJyutping\__init__.py", line 3, in <module>
    from .ToJyutping import *
  File "f:\program\anaconda3\envs\paddle.1\lib\site-packages\ToJyutping\ToJyutping.py", line 9, in <module>
    for line in f:
UnicodeDecodeError: 'gbk' codec can't decode byte 0x96 in position 2: illegal multibyte sequence

语音合成有问题

[zxcd]

文件编码问题，你看一下你的输入是否包括非法字符或者系统默认编码有问题。

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.