yime 0.1.0__py3-none-any.whl

pianyin/__init__.py

@@ -0,0 +1,4 @@
+"""pianyin 包初始化文件"""
+from .pianyin import Pianyin, PitchedPianyin, UnpitchedPianyin
+
+__all__ = ['Pianyin', 'PitchedPianyin', 'UnpitchedPianyin']

pianyin/calculate_pentatonic_value.py

@@ -0,0 +1,82 @@
+import math
+
+
+def calculate_pentatonic_value(f, f_min, f_max):
+    """
+    计算调域五度值
+    :param f: 当前频率值(Hz)
+    :param f_min: 频率最小值(Hz)
+    :param f_max: 频率最大值(Hz)
+    :return: 五度值(1-5)
+    """
+    if f_min <= 0 or f_max <= 0 or f <= 0:
+        raise ValueError("频率值必须大于0")
+    if f < f_min or f > f_max:
+        raise ValueError("当前频率必须在最小和最大频率范围内")
+
+    numerator = math.log(f) - math.log(f_min)
+    denominator = math.log(f_max) - math.log(f_min)
+    pentatonic_value = 5 * (numerator / denominator)
+
+    return pentatonic_value
+
+
+# 示例参数
+f_max = 202        # 最大频率
+f_min = 101        # 最小频率
+f_current = 202    # 当前频率
+
+# 计算五度值
+try:
+    result = calculate_pentatonic_value(f_current, f_min, f_max)
+    print(f"五度值为: {result:.2f}")
+except ValueError as e:
+    print(f"计算错误: {e}")
+
+# 示例参数
+f_max = 202        # 最大频率
+f_min = 101        # 最小频率
+f_current = 175.8    # 当前频率
+
+# 计算五度值
+try:
+    result = calculate_pentatonic_value(f_current, f_min, f_max)
+    print(f"五度值为: {result:.2f}")
+except ValueError as e:
+    print(f"计算错误: {e}")
+
+# 示例参数
+f_max = 202        # 最大频率
+f_min = 101        # 最小频率
+f_current = 153    # 当前频率
+
+# 计算五度值
+try:
+    result = calculate_pentatonic_value(f_current, f_min, f_max)
+    print(f"五度值为: {result:.2f}")
+except ValueError as e:
+    print(f"计算错误: {e}")
+
+# 示例参数
+f_max = 202        # 最大频率
+f_min = 101        # 最小频率
+f_current = 133.3    # 当前频率
+
+# 计算五度值
+try:
+    result = calculate_pentatonic_value(f_current, f_min, f_max)
+    print(f"五度值为: {result:.2f}")
+except ValueError as e:
+    print(f"计算错误: {e}")
+
+# 示例参数
+f_max = 202        # 最大频率
+f_min = 101        # 最小频率
+f_current = 116    # 当前频率
+
+# 计算五度值
+try:
+    result = calculate_pentatonic_value(f_current, f_min, f_max)
+    print(f"五度值为: {result:.2f}")
+except ValueError as e:
+    print(f"计算错误: {e}")

pianyin/calculate_speech_pentatonic_value.py

@@ -0,0 +1,64 @@
+import math
+
+
+def calculate_speech_pentatonic_value(f_current, f_min, f_max):
+    """
+    计算话语调域五度值
+    :param f_current: 当前语音频率值(Hz)
+    :param f_min: 话语调域最低频率值(Hz)
+    :param f_max: 话语调域最高频率值(Hz)
+    :return: 五度值(1-5)
+    :raises ValueError: 如果输入参数无效
+    """
+    # 验证输入参数
+    if f_min <= 0 or f_max <= 0 or f_current <= 0:
+        raise ValueError("频率值必须大于0")
+    if f_min >= f_max:
+        raise ValueError("最低频率必须小于最高频率")
+    if f_current < f_min or f_current > f_max:
+        raise ValueError("当前频率必须在调域范围内")
+
+    # 计算五度值
+    log_range = math.log(f_max) - math.log(f_min)
+    position = (math.log(f_current) - math.log(f_min)) / log_range
+    pentatonic_value = 1 + 4 * position  # 映射到1-5范围
+
+    return pentatonic_value
+
+
+def analyze_speech_range(speech_samples):
+    """
+    分析语音样本，获取调域范围
+    :param speech_samples: 语音频率样本列表(Hz)
+    :return: (f_min, f_max) 调域范围
+    :raises ValueError: 如果样本数据无效
+    """
+    if not speech_samples:
+        raise ValueError("语音样本不能为空")
+
+    return min(speech_samples), max(speech_samples)
+
+
+# 示例使用
+if __name__ == "__main__":
+    # 示例1: 已知调域范围
+    try:
+        f_min = 85   # 正常话语调域最低值示例(Hz)
+        f_max = 255  # 正常话语调域最高值示例(Hz)
+        current_pitch = 194  # 当前语音频率
+
+        result = calculate_speech_pentatonic_value(current_pitch, f_min, f_max)
+        print(f"五度值: {result:.2f}")
+    except ValueError as e:
+        print(f"计算错误: {e}")
+
+    # 示例2: 从语音样本分析
+    try:
+        speech_samples = [85, 112, 148, 194, 255]  # 实际语音频率样本
+        f_min, f_max = analyze_speech_range(speech_samples)
+
+        for sample in speech_samples:
+            result = calculate_speech_pentatonic_value(sample, f_min, f_max)
+            print(f"频率 {sample}Hz 的五度值: {result:.2f}")
+    except ValueError as e:
+        print(f"分析错误: {e}")

pianyin/calculate_speech_pentatonic_value_1.py

@@ -0,0 +1,94 @@
+import math
+
+
+def sigmoid(x, center=0.5, steepness=10):
+    """Sigmoid函数实现平滑过渡"""
+    return 1 / (1 + math.exp(-steepness * (x - center)))
+
+
+def calculate_speech_pentatonic_value(f_current, f_min, f_max, gender=None):
+    """
+    计算符合五声调式五度制映射的话语调域五度值(改进版)
+    改进点:
+    - 使用Sigmoid函数实现平滑过渡
+    - 增加性别参数适应不同基频范围
+
+    规则: 
+    - 使用对数划分，更符合五声调式的频率感知特性
+    - 通过Sigmoid函数实现五度值之间的平滑过渡
+    - 性别参数可调整划分边界(默认无差别)
+
+    :param f_current: 当前语音频率值(Hz)
+    :param f_min: 话语调域最低频率值(Hz)
+    :param f_max: 话语调域最高频率值(Hz)
+    :param gender: 性别参数('male'/'female'/None)，用于调整划分边界
+    :return: 五度值(1-5)
+    :raises ValueError: 如果输入参数无效
+    """
+    # 验证输入参数
+    if f_min <= 0 or f_max <= 0 or f_current <= 0:
+        raise ValueError("频率值必须大于0")
+    if f_min >= f_max:
+        raise ValueError("最低频率必须小于最高频率")
+    if f_current < f_min or f_current > f_max:
+        raise ValueError("当前频率必须在调域范围内")
+
+    # 计算对数位置(更符合人耳对音高的感知)
+    log_min = math.log(f_min)
+    log_max = math.log(f_max)
+    log_current = math.log(f_current)
+    position = (log_current - log_min) / (log_max - log_min)
+
+    # 根据性别调整划分边界(女声整体偏高，男声整体偏低)
+    if gender == 'female':
+        boundaries = [0.15, 0.35, 0.55, 0.75]  # 女声边界调整
+    elif gender == 'male':
+        boundaries = [0.25, 0.45, 0.65, 0.85]  # 男声边界调整
+    else:
+        boundaries = [0.20, 0.40, 0.60, 0.80]  # 默认边界
+
+    # 使用Sigmoid函数计算各度数的权重
+    w5 = sigmoid(position, boundaries[3])  # 5度权重
+    w4 = sigmoid(position, boundaries[2]) - w5  # 4度权重
+    w3 = sigmoid(position, boundaries[1]) - (w5 + w4)  # 3度权重
+    w2 = sigmoid(position, boundaries[0]) - (w5 + w4 + w3)  # 2度权重
+    w1 = 1 - (w5 + w4 + w3 + w2)  # 1度权重
+
+    # 加权计算最终五度值
+    return 1 * w1 + 2 * w2 + 3 * w3 + 4 * w4 + 5 * w5
+
+
+# 示例使用
+if __name__ == "__main__":
+    # 示例参数
+    f_min = 100  # 最低频率(Hz)
+    f_max = 400  # 最高频率(Hz)
+
+    # 测试不同频率点
+    test_frequencies = [100, 120, 150, 200, 250, 300, 350, 400]
+
+    print("默认(无性别参数):")
+    for freq in test_frequencies:
+        try:
+            result = calculate_speech_pentatonic_value(freq, f_min, f_max)
+            print(f"频率 {freq}Hz 的五度值: {result:.2f}")
+        except ValueError as e:
+            print(f"频率 {freq}Hz 计算错误: {e}")
+
+    print("\n女声:")
+    for freq in test_frequencies:
+        try:
+            result = calculate_speech_pentatonic_value(
+                freq, f_min, f_max, 'female')
+            print(f"频率 {freq}Hz 的五度值: {result:.2f}")
+        except ValueError as e:
+            print(f"频率 {freq}Hz 计算错误: {e}")
+
+    print("\n男声:")
+    for freq in test_frequencies:
+        try:
+            result = calculate_speech_pentatonic_value(
+                freq, f_min, f_max, 'male')
+            print(f"频率 {freq}Hz 的五度值: {result:.2f}")
+        except ValueError as e:
+            print(f"频率 {freq}Hz 计算错误: {e}")

pianyin/calculate_tone_median.py

@@ -0,0 +1,100 @@
+class ToneMedianCalculator:
+    """
+    五度标调法调域中值计算器
+    1. 计调单个调值中值：
+    2. 计算单个调类中值：
+    3. 计算整个调系中值：
+    """
+
+    @staticmethod
+    def calculate_tone_value_median(pitch_levels):
+        """
+        计算单个调值中值
+        :param pitch_levels: 调值列表，如[5,5,5]代表阴平
+        :return: 调值中值
+        """
+        if not pitch_levels:
+            raise ValueError("调值列表不能为空")
+        return sum(pitch_levels) / len(pitch_levels)
+
+    @staticmethod
+    def calculate_tone_category_median(tone_values_medians):
+        """
+        计算单个调类中值
+        :param tone_values_medians: 该调类的多个调值中值列表
+        :return: 调类中值
+        """
+        if not tone_values_medians:
+            raise ValueError("调值中值列表不能为空")
+        return sum(tone_values_medians) / len(tone_values_medians)
+
+    @staticmethod
+    def calculate_tone_system_median(category_medians):
+        """
+        计算整个调系中值
+        :param category_medians: 各调类中值列表
+        :return: 调系中值
+        """
+        if not category_medians:
+            raise ValueError("调类中值列表不能为空")
+        return sum(category_medians) / len(category_medians)
+
+    @staticmethod
+    def calculate_variation_coefficient(values):
+        """计算变异系数（标准差/均值）评估离散程度"""
+        mean = sum(values) / len(values)
+        std_dev = (sum((x - mean)**2 for x in values) / len(values))**0.5
+        return std_dev / mean
+
+    # 应用示例
+    low_tone_vc = calculate_variation_coefficient(
+        [1.33, 4.00, 2.33])  # 上声变异系数
+    falling_tone_vc = calculate_variation_coefficient([3.33, 4.00])  # 去声变异系数
+
+
+# 示例使用
+if __name__ == "__main__":
+    # 1. 计算各调值中值
+    high_tone = ToneMedianCalculator.calculate_tone_value_median([5, 5, 5])  # 阴平
+    rising_tone = ToneMedianCalculator.calculate_tone_value_median([
+                                                                3, 4, 5])  # 阳平
+    low_tone1 = ToneMedianCalculator.calculate_tone_value_median([
+                                                                   2, 1, 1])  # 上声1
+    low_tone2 = ToneMedianCalculator.calculate_tone_value_median([
+                                                                   3, 4, 5])  # 上声2
+    low_tone3 = ToneMedianCalculator.calculate_tone_value_median([
+                                                                   2, 1, 4])  # 上声3
+    falling_tone1 = ToneMedianCalculator.calculate_tone_value_median([
+                                                                5, 4, 1])  # 去声1
+    falling_tone2 = ToneMedianCalculator.calculate_tone_value_median([
+                                                                5, 4, 3])  # 去声2
+
+    print(f"阴平调值中值: {high_tone:.2f}")
+    print(f"阳平调值中值: {rising_tone:.2f}")
+    print(f"上声1调值中值: {low_tone1:.2f}")
+    print(f"上声2调值中值: {low_tone2:.2f}")
+    print(f"上声3调值中值: {low_tone3:.2f}")
+    print(f"去声1调值中值: {falling_tone1:.2f}")
+    print(f"去声2调值中值: {falling_tone2:.2f}")
+
+    # 2. 计算各调类中值
+    high_tone_median = ToneMedianCalculator.calculate_tone_category_median([
+                                                                         high_tone])
+    rising_tone_median = ToneMedianCalculator.calculate_tone_category_median([
+                                                                          rising_tone])
+    low_tone_median = ToneMedianCalculator.calculate_tone_category_median(
+        [low_tone1, low_tone2, low_tone3])
+    falling_tone_median = ToneMedianCalculator.calculate_tone_category_median([
+                                                                         falling_tone1, falling_tone2])
+
+    print(f"\n阴平调类中值: {high_tone_median:.2f}")
+    print(f"阳平调类中值: {rising_tone_median:.2f}")
+    print(f"上声调类中值: {low_tone_median:.2f}")
+    print(f"去声调类中值: {falling_tone_median:.2f}")
+
+    # 3. 计算调系中值
+    system_median = ToneMedianCalculator.calculate_tone_system_median([
+        high_tone_median, rising_tone_median, low_tone_median, falling_tone_median
+    ])
+
+    print(f"\n调系调域中值: {system_median:.2f}")

pianyin/indeterminate_pitch_pianyin.py

@@ -0,0 +1,64 @@
+# 噪音类片音表示法
+# 噪音类片音分为两类：清辅音和浊辅音。
+# 清音没有音调，浊辅音类噪音有非规律性的音高特征。
+# 浊辅音类噪音的音调不具有辨义作用，通常不标调。
+# 在通用现代汉语中，噪音类片音就是声母的音值（分布在音节中的声母的实际发音）。
+# 噪音类表示汉语音节的噪音
+
+from abc import ABC, abstractmethod
+from typing import Optional
+
+
+class UnpitchedPianyin(ABC):
+    """噪音类片音，清音无音调，浊辅音可能有音调"""
+
+    def __init__(self, quality: str, duration: str = '', loudness: str = '', pitch: Optional[str] = None):
+        """
+        初始化噪音对象
+        :param quality: 音质(必选)
+        :param duration: 音长(默认)
+        :param loudness: 音强(默认)
+        :param pitch: 音调(可选)，None表示清音无音调，不传或特定值表示浊辅音可能有音调
+        """
+        self.quality = quality
+        self.duration = duration
+        self.loudness = loudness
+        self.pitch = pitch
+
+    def clear_pitch(self) -> None:
+        """明确定义为清音(无音调)"""
+        self.pitch = None
+
+    @abstractmethod
+    def is_valid(self) -> bool:
+        """检查噪音对象是否有效，子类必须实现"""
+        pass
+
+    def __str__(self) -> str:
+        """返回噪音的通用字符串表示"""
+        return (f"{self.__class__.__name__}(quality='{self.quality}'"
+                f"{f', pitch={repr(self.pitch)}' if self.pitch is not None else ''}"
+                f"{f', duration={repr(self.duration)}' if self.duration else ''}"
+                f"{f', loudness={repr(self.loudness)}' if self.loudness else ''})")
+
+
+class ClearPianyin(UnpitchedPianyin):
+    """清音类噪音，无音调"""
+
+    def __init__(self, quality: str, duration: str = '', loudness: str = ''):
+        super().__init__(quality=quality, duration=duration, loudness=loudness, pitch=None)
+
+    def is_valid(self) -> bool:
+        """清音只需要有音质即可有效"""
+        return bool(self.quality)
+
+
+class VoicedUnpitchedPianyin(UnpitchedPianyin):
+    """浊辅音类噪音，可能有音调"""
+
+    def __init__(self, quality: str, pitch: Optional[str] = None, duration: str = '', loudness: str = ''):
+        super().__init__(quality=quality, pitch=pitch, duration=duration, loudness=loudness)
+
+    def is_valid(self) -> bool:
+        """浊辅音需要有音质，音调可选"""
+        return bool(self.quality)

pianyin/initial.py

@@ -0,0 +1,112 @@
+# Initial Sound Representation Module
+# 首音表示模块
+#
+# This module provides a class to represent the initial sound in Chinese phonetics,
+# supporting multiple representation methods including pinyin, phonetic, pianyin and yinyuan.
+#
+# 首音: 由声母和与声母联结的调段构成的音段
+# 声母 (initial): 必选属性
+# 首调: 与声母联结的调段, 可选属性, 默认值 None
+#
+# The initial sound is a syllable segment composed of the initial consonant and the tone segment connected to the initial consonant
+# The initial consonant is a required attribute
+# The tone of the initial sound (the tone segment connected to the initial consonant), named initial tone, is an optional attribute
+
+from pianyin.pianyin import UnpitchedPianyin  # Ensure pianyin/pianyin.py exists and contains UnpitchedPianyin
+
+
+class Initial(UnpitchedPianyin):
+    """
+    Represents the initial sound of a Chinese syllable.
+    表示汉语音节的首音
+
+    Attributes:
+        quality (str): The quality of the initial consonant (声母/initial)
+        tone_segment (str|None): The tone segment using 5-level notation (0-5)
+        representation (str): Representation method (pinyin/phonetic/pianyin/yinyuan)
+        pitch_style (str): Tone display style ('number' or 'mark')
+    """
+    # 以下内容保持不变...
+    TONE_SEGMENT_MARKS = {
+        "5": "˥",  # 高平
+        "4": "˦",  # 次高平调
+        "3": "˧",  # 中平调
+        "2": "˨",  # 次低平调
+        "1": "˩",  # 低调
+        "0": "꜌"   # 与清声母联结的调段
+    }
+
+    def __init__(self, consonant, tone_segment=None, representation='pinyin', pitch_style='number'):
+        """
+        Initialize an Initial instance.
+        初始化首音实例
+
+        Args:
+            consonant (str): The initial consonant (声母/initial)
+            tone_segment (str|None): The tone segment (0-5) or None for no tone segment (与清声母联结的调段)
+            representation (str): Representation method (default 'pinyin')
+            pitch_style (str): Tone display style ('number' or 'mark', default 'number')
+        """
+        super().__init__(quality=consonant)
+        self.tone_segment = str(
+            tone_segment) if tone_segment is not None else None
+        self.representation = representation.lower()
+        self.pitch_style = pitch_style.lower()
+
+        self._validate_input()
+
+    def _validate_input(self):
+        """Validate the input parameters. 验证输入参数"""
+        if not isinstance(self.quality, str) or not self.quality:
+            raise ValueError("Consonant must be a non-empty string")
+
+        if self.tone_segment is not None and self.tone_segment not in ["0", "1", "2", "3", "4", "5"]:
+            raise ValueError("Tone segment must be between 0-5 or None")
+
+        if self.representation not in ['pinyin', 'phonetic', 'pianyin', 'yinyuan']:
+            raise ValueError("Invalid representation method")
+
+        if self.pitch_style not in ['number', 'mark']:
+            raise ValueError("Tone style must be 'number' or 'mark'")
+
+    def _get_tone_mark(self):
+        """Get tone mark for current tone segment. 获取当前调段的符号标调"""
+        if self.tone_segment is None:
+            return ""
+        return self.TONE_SEGMENT_MARKS.get(self.tone_segment, "")
+
+    def __repr__(self):
+        """General string representation. 一般字符串表示"""
+        return f"Initial(quality={self.quality!r}, tone_segment={self.tone_segment!r}, representation={self.representation!r}, pitch_style={self.pitch_style!r})"
+
+    def __str__(self):
+        """简式字符串表示. 只包含必选项 consonant"""
+        return f"{self.quality}"
+
+    @classmethod
+    def generate_from_consonant_table(cls, consonant_table, representation='pinyin', pitch_style='number'):
+        """
+        从声母表生成首音对象映射
+        Args:
+            consonant_table (list/dict): 声母列表或字典
+            representation (str): 表示方法
+            pitch_style (str): pitch_style
+        Returns:
+            dict: 声母到首音对象的映射
+        """
+        if isinstance(consonant_table, list):
+            return {consonant: cls(consonant, representation=representation, pitch_style=pitch_style)
+                    for consonant in consonant_table}
+        elif isinstance(consonant_table, dict):
+            # 处理嵌套字典结构，只使用键(拼音字母)作为声母
+            return {k: cls(k, representation=representation, pitch_style=pitch_style)
+                    for k in consonant_table.keys()}
+        else:
+            raise ValueError("consonant_table 必须是列表或字典")
+
+    def is_valid(self) -> bool:
+        """检查首音对象是否有效"""
+        return bool(self.quality) and (
+            self.tone_segment is None or
+            self.tone_segment in ["0", "1", "2", "3", "4", "5"]
+        )

pianyin/pianyin.py

@@ -0,0 +1,66 @@
+# 片音(Pianyin)表示法
+# 片音具有音质、音调、音长和音强四个属性
+# 在通用现代汉语中，片音的音质和音调是必选的，音长和音强是默认的
+# 片音类表示汉语音节的片音
+# Pianyin notation
+# A Pianyin has four attributes: quality, pitch, duration, and loudness
+# In Standard Modern Chinese, quality and pitch are required, while duration and loudness are default
+# The Pianyin class represents the phonetic components of a Chinese syllable
+# 片音(Pianyin)分成噪音和乐音两类
+# 噪音的音调是空，音质就是噪音
+# 乐音的音调非空，音质和音调构成乐音
+
+from abc import ABC, abstractmethod
+from typing import Optional
+
+
+class Pianyin(ABC):
+    """片音(Pianyin)抽象基类，表示汉语音节的片音"""
+
+    def __init__(self, quality: str, pitch: Optional[str] = None,
+                 duration: str = 'neutral', loudness: str = 'neutral'):
+        """
+        初始化片音对象
+        :param quality: 音质(必选)
+        :param pitch: 音调(乐音必选，噪音可选)
+        :param duration: 音长(默认)
+        :param loudness: 音强(默认)
+        """
+        self.quality = quality
+        self.pitch = pitch
+        self.duration = duration
+        self.loudness = loudness
+
+    @abstractmethod
+    def is_valid(self) -> bool:
+        """检查片音对象是否有效，子类必须实现"""
+        pass
+
+    def __str__(self) -> str:
+        """返回片音的通用字符串表示"""
+        return (f"{self.__class__.__name__}(quality='{self.quality}'"
+                f"{f', pitch={repr(self.pitch)}' if self.pitch is not None else ''}"
+                f"{f', duration={repr(self.duration)}' if self.duration else ''}"
+                f"{f', loudness={repr(self.loudness)}' if self.loudness else ''})")
+
+
+class UnpitchedPianyin(Pianyin):
+    """噪音类片音，音调为空，音质为噪音"""
+
+    def __init__(self, quality: str, duration: str = 'neutral', loudness: str = 'neutral'):
+        super().__init__(quality=quality, pitch=None, duration=duration, loudness=loudness)
+
+    def is_valid(self) -> bool:
+        """噪音只需要有音质即可有效"""
+        return bool(self.quality)
+
+
+class PitchedPianyin(Pianyin):
+    """乐音类片音，音调非空，音质和音调构成乐音"""
+
+    def __init__(self, quality: str, pitch: str, duration: str = 'neutral', loudness: str = 'neutral'):
+        super().__init__(quality=quality, pitch=pitch, duration=duration, loudness=loudness)
+
+    def is_valid(self) -> bool:
+        """乐音需要有音质和音调才能有效"""
+        return bool(self.quality and self.pitch)