【记录】Python3｜用百度语音 API 朗读你的小说TXT

百度语音合成官方教程_AI开放平台

百度语音合成官方demo_github.com

简单地写了一个按段落朗读文本的demo：DEMO链接_gitee.com。
有时候会请求不到数据，不知道是网络原因还是什么，已添加自动重新请求。

config.ini：

;关于语音合成的相关配置
[default]
api_key = Your api key
secret_key = Your secret key
;发音人选择, 基础音库：0为度小美，1为度小宇，3为度逍遥，4为度丫丫，
;精品音库：5为度小娇，103为度米朵，106为度博文，110为度小童，111为度小萌，默认为度小美
per = 3
;语速，取值0-15，默认为5中语速
spd = 4
;音调，取值0-15，默认为5中语调
pit = 5
;音量，取值0-9，默认为5中音量
vol = 5
# 下载的文件格式, 3：mp3(default) 4： pcm-16k 5： pcm-8k 6. wav
aue = 3
;下载的文件格式, 可选项：mp3(default), pcm-16k, pcm-8k, wav
format = mp3
cuid = 123456PYTHON
tts_url = http://tsn.baidu.com/text2audio
token_url = http://openapi.baidu.com/oauth/2.0/token
;有此scope表示有tts能力，没有请在网页里勾选
scope = audio_tts_post

[追风筝的人.txt]
text_lct = 12

main.py

# coding=utf-8
import os
import json
from configparser import ConfigParser
from playsound import playsound

from urllib.request import urlopen
from urllib.request import Request
from urllib.error import URLError
from urllib.error import HTTPError
from urllib.parse import urlencode
from urllib.parse import quote_plus

TEXT = "欢迎使用百度语音合成。"
ini_file = "./config.ini"
cfg_name = "default"
book = "D:/总要删的/追风筝的人.txt"


def load_config(ini, name):
    cfg = ConfigParser()
    # 读取文件内容
    cfg.read(ini, encoding="gbk")
    # cfg.items()返回list，元素为tuple
    return dict(cfg.items(name))


class DemoError(Exception):
    pass


def fetch_token(dft_cfg):
    # print("fetch token begin")
    params = {'grant_type': 'client_credentials',
              'client_id': dft_cfg['api_key'],
              'client_secret': dft_cfg['secret_key']}
    post_data = urlencode(params)
    post_data = post_data.encode('utf-8')
    req = Request(dft_cfg['token_url'], post_data)
    try:
        f = urlopen(req, timeout=5)
        result_str = f.read()
    except URLError as err:
        print('token http response http code : ' + str(err.code))
        result_str = err.read()
    result_str = result_str.decode()

    # print(result_str)
    result = json.loads(result_str)
    # print(result)
    if 'access_token' in result.keys() and 'scope' in result.keys():
        if not dft_cfg['scope'] in result['scope'].split(' '):
            raise DemoError('scope is not correct')
        # print('SUCCESS WITH TOKEN: %s ; EXPIRES IN SECONDS: %s' % (result['access_token'], result['expires_in']))
        return result['access_token']
    else:
        raise DemoError('MAYBE API_KEY or SECRET_KEY not correct: access_token or scope not found in token response')


def update_text(file, book_title, ini):
    # 读取配置文件
    cfg = ConfigParser()
    # 读取文件内容
    cfg.read(ini, encoding="gbk")
    if cfg.has_option(book_title, "text_lct"):
        now_lct = int(cfg.get(book_title, "text_lct"))
    else:
        cfg.add_section(book_title)
        now_lct = 0

    if len(file) <= now_lct:
        return "已经读到最后一句啦！换本书吧~！"
    else:
        while not len(file[now_lct].strip()):
            now_lct = now_lct + 1
        # 更新配置文件
        cfg.set(book_title, "text_lct", str(now_lct + 1))
        cfg.write(open(ini, "r+"))
        return file[now_lct]


def request_api(params):
    data = urlencode(params)
    req = Request(dft_cfg['tts_url'], data.encode('utf-8'))
    try:
        f = urlopen(req)
        result_str = f.read()
        headers = dict((name.lower(), value) for name, value in f.headers.items())
        has_error = ('content-type' not in headers.keys() or headers['content-type'].find('audio/') < 0)
    except Exception as e:
        print('asr http response http code : ' + str(e))
        result_str = str(e)
        has_error = True
    if has_error:
        print("tts api  error:" + str(result_str, 'utf-8'))
        request_api(params)
    else:
        # Step 3.4: 保存请求的音频结果并输出成temp.mp3，朗读完毕后删除
        save_file = "error.txt" if has_error else 'temp.' + dft_cfg['format']
        with open(save_file, 'wb') as of:
            of.write(result_str)
        playsound(save_file)
        os.remove(save_file)


if __name__ == '__main__':
    # Step 1: 载入配置文件
    dft_cfg = load_config(ini_file, cfg_name)
    # Step 2: 获取Token
    token = fetch_token(dft_cfg)
    # Step 3: 向API发起请求
    # Step 3.1: 初始化请求参数params、书籍标题
    params = {'tok': token, 'tex': '', 'per': dft_cfg['per'], 'spd': dft_cfg['spd'], 'pit': dft_cfg['pit'],
              'vol': dft_cfg['vol'], 'aue': dft_cfg['aue'], 'cuid': dft_cfg['cuid'],
              'lan': 'zh', 'ctp': 1}  # lan ctp 固定参数
    book_title = (book.split('/'))[-1]
    # 打开指定书籍, 并按行读取
    with open(book, "r", encoding='utf-8') as f:
        file = f.readlines()
    # Step 3.2: 不断获取文本并朗读请求得到的音频
    while 1:
        # Step 3.2.1: 根据上次阅读的位置，更新需要合成的文本内容
        TEXT = update_text(file, book_title, ini_file)
        print(TEXT)
        params['tex'] = quote_plus(TEXT)  # 此处TEXT需要两次urlencode
        # Step 3.2.2: 将参数打包，并向指定URL请求，并朗读
        request_api(params)

目前的结果：