model.py•6.31 kB
# Copyright (c) 2023 Baidu, Inc. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
r"""short text to speech model."""
import proto
from pydantic import BaseModel
class TTSRequest(proto.Message):
r"""文本转语音请求参数.
属性:
tex (str): 如果选择`baidu-tts`模型，`tex`最大文本长度为1024 GBK编码长度, 如果选择`paddlespeech-tts`模型,
`tex`最大文本长度是510个字符.
tok (str, 可选): 用户token.
cuid (str, 可选): 用户ID.
ctp (str, 可选): 用户客户端类型.
lan (str): 目前仅支持中英-ZH.
spd(int, 可选): 语音语速，默认是5中等语速，取值范围在0~15之间，如果选择模型为paddlespeech-tts，参数自动失效.
pit(int, 可选): 语音音调，默认是5中等音调，取值范围在0~15之间，如果选择模型为paddlespeech-tts，参数自动失效.
vol(int, 音量): 语音音量，默认是5中等音量，取值范围在0~15之间，如果选择模型为paddlespeech-tts，参数自动失效.
per(int, 可选): 语音人物特征，默认是0,可选值包括度小宇=1 度小美=0 度逍遥（基础）=3 度丫丫=4 度逍遥（精品）=5003
度小鹿=5118 度博文=106 度小童=110 度小萌=111 度米朵=103 度小娇=5
度逍遥-情感男声=4003 度博文-专业男主播=4106 度小贤-电台男主播=4115
度小鹿-甜美女声=4119 度灵儿-清激女声=4105 度小乔-活泼女声=4117
度小雯-活力女主播=4100 度米朵-可爱女声=4103 度姗姗-娱乐女声=4144
度小贝-知识女主播=4278 度清风-配音男声=4143 度小新-专业女主播=4140
度小彦-知识男主播=4129 度星河-广告男声=4149 度小清-广告女声=4254
度博文-综艺男声=4206 南方-电台女主播=4226，
如果选择模型为paddlespeech-tts，参数自动失效.
aue(int, 可选): 语音格式, 默认是3(mp3) 4(pcm-16k) 5(pcm-8k) 6-wav.
tp_project_id(str): paddlespeech-tts项目ID
tp_per_id(str): paddlespeech-tts音频ID
"""
tex: str = proto.Field(
proto.STRING,
number=1,
)
tok: str = proto.Field(
proto.STRING,
number=2,
optional=True,
)
cuid: str = proto.Field(
proto.STRING,
number=3,
optional=True,
)
ctp: str = proto.Field(
proto.STRING,
number=4,
)
lan: str = proto.Field(
proto.STRING,
number=5,
)
spd: int = proto.Field(
proto.INT32,
number=6,
)
pit: int = proto.Field(
proto.INT32,
number=7,
)
vol: int = proto.Field(
proto.INT32,
number=8,
)
per: int = proto.Field(
proto.INT32,
number=9,
)
aue: int = proto.Field(
proto.INT32,
number=10,
)
tp_project_id: str = proto.Field(
proto.STRING,
number=11,
optional=True,
)
tp_per_id: str = proto.Field(
proto.STRING,
number=12,
optional=True,
)
stream: bool = proto.Field(
proto.BOOL,
number=13,
optional=True,
)
def __validate(self):
"""检查公共参数"""
if len(self.tex) == 0:
raise ValueError("tex field must be set")
if self.spd < 0 or self.spd > 15:
raise ValueError("spd value must in [0,15]")
if self.pit < 0 or self.pit > 15:
raise ValueError("pit value must in [0,15]")
if self.vol < 0 or self.vol > 15:
raise ValueError("vol value must in [0,15]")
def validate_baidu_tts(self):
"""检查baidu-tts模型请求参数"""
self.__validate()
_BAIDU_VALID_PER = {0, 1, 3, 4, 5, 103, 106, 110, 111, 4003, 4106, 4115, 4119,
4105, 4117, 4100, 4103, 4144, 4278, 4143, 4140, 4129, 4149, 4254, 4206, 4226, 5003, 5118}
if self.per not in _BAIDU_VALID_PER:
raise ValueError(
f"per value is illegal, expected in {_BAIDU_VALID_PER}, got {self.per}"
)
if self.aue == 0:
self.aue = 3
if self.aue not in {3, 4, 5, 6}:
raise ValueError(f"aue value is illegal, exepect in {3, 4, 5, 6}, got {self.aue}")
def validate_paddle_speech_tts(self):
"""验证参数是否合法"""
self.__validate()
if self.aue == 0:
self.aue = 6
if self.aue not in {6}:
raise ValueError("aue value is illegal, expectd aue value is 6, got{}".format(self.aue))
class TTSResponse(proto.Message):
r"""文本转语音返回.
属性:
binary (bytes): 语音二进制流.
aue (int):语音格式, 3(mp3), 4(pcm-16k), 5(pcm-8k) 6(wav).
request_id(str): 请求ID
"""
binary: bytes = proto.Field(
proto.BYTES,
number=1
)
aue: int = proto.Field(
proto.INT32,
number=2,
)
request_id: str = proto.Field(
proto.STRING,
number=3,
)
class TTSInMsg(BaseModel):
r"""文本转语音输入消息.
属性:
text(str): 待转为语音的文本, 如果选择`baidu-tts`模型，最大文本长度为1024 GBK编码长度
如果选择`paddlespeech-tts`模型, 最大文本长度是510个字符.
"""
text: str
class TTSOutMsg(BaseModel):
r""" 文本转语音输出消息.
属性:
audio_binary(bytes): 语音二进制流.
audio_type(AudioType): 语音类型，`mp3`或`wav`.
"""
audio_binary: bytes
audio_type: str