feat(azure_voice): add language detection, support mulitple languages

requirements-optional.txt

@@ -7,6 +7,7 @@ gTTS>=2.3.1 # google text to speech
 pyttsx3>=2.90 # pytsx text to speech
 baidu_aip>=4.16.10 # baidu voice
 # azure-cognitiveservices-speech # azure voice
+langid # language detect
 
 #install plugin
 dulwich

translate/baidu/baidu_translate.py

@@ -8,8 +8,6 @@ import requests
 from config import conf
 from translate.translator import Translator
 
-# from langid import classify
-
 
 class BaiduTranslator(Translator):
     def __init__(self) -> None:
@@ -24,7 +22,6 @@ class BaiduTranslator(Translator):
     def translate(self, query: str, from_lang: str = "", to_lang: str = "en") -> str:
         if not from_lang:
             from_lang = "auto"  # baidu suppport auto detect
-            # from_lang = classify(query)[0]
         salt = random.randint(32768, 65536)
         sign = self.make_md5(self.appid + query + str(salt) + self.appkey)
         headers = {"Content-Type": "application/x-www-form-urlencoded"}

voice/azure/azure_voice.py

@@ -6,6 +6,7 @@ import os
 import time
 
 import azure.cognitiveservices.speech as speechsdk
+from langid import classify
 
 from bridge.reply import Reply, ReplyType
 from common.log import logger
@@ -30,7 +31,15 @@ class AzureVoice(Voice):
             config = None
             if not os.path.exists(config_path):  # 如果没有配置文件，创建本地配置文件
                 config = {
-                    "speech_synthesis_voice_name": "zh-CN-XiaoxiaoNeural",
+                    "speech_synthesis_voice_name": "zh-CN-XiaoxiaoNeural",  # 识别不出时的默认语音
+                    "auto_detect": True,  # 是否自动检测语言
+                    "speech_synthesis_zh": "zh-CN-XiaozhenNeural",
+                    "speech_synthesis_en": "en-US-JacobNeural",
+                    "speech_synthesis_ja": "ja-JP-AoiNeural",
+                    "speech_synthesis_ko": "ko-KR-SoonBokNeural",
+                    "speech_synthesis_de": "de-DE-LouisaNeural",
+                    "speech_synthesis_fr": "fr-FR-BrigitteNeural",
+                    "speech_synthesis_es": "es-ES-LaiaNeural",
                     "speech_recognition_language": "zh-CN",
                 }
                 with open(config_path, "w") as fw:
@@ -38,11 +47,12 @@ class AzureVoice(Voice):
             else:
                 with open(config_path, "r") as fr:
                     config = json.load(fr)
+            self.config = config
             self.api_key = conf().get("azure_voice_api_key")
             self.api_region = conf().get("azure_voice_region")
             self.speech_config = speechsdk.SpeechConfig(subscription=self.api_key, region=self.api_region)
-            self.speech_config.speech_synthesis_voice_name = config["speech_synthesis_voice_name"]
+            self.speech_config.speech_synthesis_voice_name = self.config["speech_synthesis_voice_name"]
-            self.speech_config.speech_recognition_language = config["speech_recognition_language"]
+            self.speech_config.speech_recognition_language = self.config["speech_recognition_language"]
         except Exception as e:
             logger.warn("AzureVoice init failed: %s, ignore " % e)
 
@@ -59,6 +69,16 @@ class AzureVoice(Voice):
         return reply
 
     def textToVoice(self, text):
+        if self.config.get("auto_detect"):
+            lang = classify(text)[0]
+            key = "speech_synthesis_" + lang
+            if key in self.config:
+                logger.info("[Azure] textToVoice auto detect language={}, voice={}".format(lang, self.config[key]))
+                self.speech_config.speech_synthesis_voice_name = self.config[key]
+            else:
+                self.speech_config.speech_synthesis_voice_name = self.config["speech_synthesis_voice_name"]
+        else:
+            self.speech_config.speech_synthesis_voice_name = self.config["speech_synthesis_voice_name"]
         fileName = TmpDir().path() + "reply-" + str(int(time.time())) + ".wav"
         audio_config = speechsdk.AudioConfig(filename=fileName)
         speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=self.speech_config, audio_config=audio_config)

voice/azure/config.json.template

@@ -1,4 +1,12 @@
 {
   "speech_synthesis_voice_name": "zh-CN-XiaoxiaoNeural",
+  "auto_detect": true,
+  "speech_synthesis_zh": "zh-CN-YunxiNeural",
+  "speech_synthesis_en": "en-US-JacobNeural",
+  "speech_synthesis_ja": "ja-JP-AoiNeural",
+  "speech_synthesis_ko": "ko-KR-SoonBokNeural",
+  "speech_synthesis_de": "de-DE-LouisaNeural",
+  "speech_synthesis_fr": "fr-FR-BrigitteNeural",
+  "speech_synthesis_es": "es-ES-LaiaNeural",
   "speech_recognition_language": "zh-CN"
 }