Question

我在运行下面的代码时遇到问题，在t one_freq = tone_map [tone_name ”行，它会导致错误，即列表索引必须是整数或切片而不是字符串

评论的音调网站上的Json示例是

[
 {
   "432": "Note",
   "434": "Frequency (Hz)",
   "436": "Wavelength (cm)",
   "438": "",
   "440": "",
   "442": "",
   "444": "",
   "446": ""
  },
 {
   "432": "C0",
   "434": "16.35",
   "436": "2109.89",
   "438": "",
   "440": "",
   "442": "",
   "444": "",
   "446": ""
  },
 {
   "432": "C#0/Db0",
   "434": "17.32",
   "436": "1991.47",
   "438": "",
   "440": "",
   "442": "",
   "444": "",
   "446": ""
  },

我用来合成音调以生成音乐的python代码是这个。我已将http://www.phy.mtu.edu/~suits/notefreqs.html处的html表转换为正在使用的json文件，但说它必须是整数或切片

import json

import numpy as np
import matplotlib.pyplot as plt
from scipy.io.wavfile import write

# Synthesize the tone based on the input parameters
def tone_synthesizer(freq, duration, amplitude=1.0, sampling_freq=44100):
# Construct the time axis 
    time_axis = np.linspace(0, duration, duration * sampling_freq)

# Construct the audio signal
    signal = amplitude * np.sin(2 * np.pi * freq * time_axis)

    return signal.astype(np.int16) 

if __name__=='__main__':
    # Names of output files
    file_tone_single = 'generated_tone_single.wav'
    file_tone_sequence = 'generated_tone_sequence.wav'

    # Source: http://www.phy.mtu.edu/~suits/notefreqs.html
    mapping_file = 'tone_mapping.json'

   # Load the tone to frequency map from the mapping file
with open(mapping_file, 'r') as f:
    tone_map = json.loads(f.read())

# Set input parameters to generate 'F' tone
tone_name = 'F'
duration = 3     # seconds
amplitude = 12000
sampling_freq = 44100    # Hz

# Extract the tone frequency
tone_freq = tone_map[tone_name]

# Generate the tone using the above parameters
synthesized_tone = tone_synthesizer(tone_freq, duration, amplitude, sampling_freq)

# Write the audio signal to the output file
write(file_tone_single, sampling_freq, synthesized_tone)

# Define the tone sequence along with corresponding durations in seconds
tone_sequence = [('G', 0.4), ('D', 0.5), ('F', 0.3), ('C', 0.6), ('A', 0.4)]

# Construct the audio signal based on the above sequence 
signal = np.array([])
for item in tone_sequence:
    # Get the name of the tone 
    tone_name = item[0]

    # Extract the corresponding frequency of the tone
    freq = tone_map[tone_name]

    # Extract the duration
    duration = item[1]

    # Synthesize the tone
    synthesized_tone = tone_synthesizer(freq, duration, amplitude, sampling_freq)

    # Append the output signal
    signal = np.append(signal, synthesized_tone, axis=0)

# Save the audio in the output file
write(file_tone_sequence, sampling_freq, signal)

Answer 1

来自python文档：

json.loads（s [，encoding [，cls [，object_hook [，parse_float [，parse_int [，parse_constant [，object_pairs_hook [，** kw]]]]]]]]]]]））

使用此转换表将s（包含JSON文档的str或unicode实例）反序列化为Python对象。

如果s是一个str实例，并且使用UTF-8（例如latin-1）以外的基于ASCII的编码进行编码，则必须指定适当的编码名称。不允许使用非基于ASCII的编码（例如UCS-2），并且应先将其解码为unicode。

从错误中您可以了解，tone_map是一个列表。试试这个：

for i in tone_map: tone_freq = i[tone_name]

将tone_freq存储在列表中并使用它

tone_frequencies = list() for i in tone_map: tone_frequencies.append(i)

Answer 2

无论您将HTML表（http://www.phy.mtu.edu/~suits/notefreqs.html）转换为JSON的方式是什么，似乎都是以这种格式为您带来了极大的不便。如果查看JSON文档，则它是一个数组（不是map / dict）。数组元素似乎与原始表行相对应，每个数组元素本身都是字典，包含键“ 432”，“ 434”，“ 436”，“ 438”等。其中的前三个是表的三列（名称，频率，波长），其余为空。

您首先应该尝试将其重新设置为更有用的格式，例如，音符名称和频率之间的映射（不要忘记将频率数据转换为浮点数-JSON文档将其作为字符串）。也许像这样的事情会做：

name_column="432"
freq_column="434"
fmap = { row[name_column] : float(row[freq_column]) for row in json_doc[1:] }

（请注意，我已经切掉了第一个数组元素，其中包含列名）

现在，您可以尝试：

freq = fmap["E0"]
print (freq)

此外，在示例Python代码中，您使用的注释名称如“ C”，但在表中找不到该名称（它具有“ C0”，“ C1”等）。

json文件只能采用整数或切片格式，不能采用字符串

2 个答案: