killua/TakwayPlatform
1
0 Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update: 在tts_info中加入了speed字段

Browse Source
This commit is contained in:
killua committed 2024-05-23 10:05:10 +08:00
1 parent 387c277c28
commit b3e0f26937
2 files changed
+16 -15
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

app/controllers/chat_controller.py
+9 -8
View File
@@ -148,7 +148,8 @@ async def create_chat_handler(chat: ChatCreateRequest, db, redis):
"speaker_id":db_character.voice_id,
"noise_scale": 0.1,
"noise_scale_w":0.668,
"length_scale": 1.2
"length_scale": 1.2,
"speend":1
}
llm_info = {
"model": "abab5.5-chat",
@@ -318,12 +319,12 @@ async def sct_llm_handler(ws,session_id,response_type,llm_info,tts_info,db,redis
audio,sr = tts._base_tts(text=sentence,
noise_scale=tts_info["noise_scale"],
noise_scale_w=tts_info["noise_scale_w"],
speed=tts_info["length_scale"])
speed=tts_info["speed"])
else:
audio,sr = tts.synthesize(text=sentence,
noise_scale=tts_info["noise_scale"],
noise_scale_w=tts_info["noise_scale_w"],
speed=tts_info["length_scale"],
speed=tts_info["speed"],
target_se=target_se)
response_message = {"type": "text", "code":200, "msg": sentence}
await ws.send_bytes(audio) #返回音频数据
@@ -499,12 +500,12 @@ async def scl_llm_handler(ws,session_id,response_type,llm_info,tts_info,db,redis
audio,sr = tts._base_tts(text=sentence,
noise_scale=tts_info["noise_scale"],
noise_scale_w=tts_info["noise_scale_w"],
speed=tts_info["length_scale"])
speed=tts_info["speed"])
else:
audio,sr = tts.synthesize(text=sentence,
noise_scale=tts_info["noise_scale"],
noise_scale_w=tts_info["noise_scale_w"],
speed=tts_info["length_scale"],
speed=tts_info["speed"],
target_se=target_se)
response_message = {"type": "text", "code":200, "msg": sentence}
await ws.send_bytes(audio)
@@ -683,17 +684,17 @@ async def voice_call_llm_handler(ws,session_id,llm_info,tts_info,db,redis,asr_re
audio,sr = tts._base_tts(text=sentence,
noise_scale=tts_info["noise_scale"],
noise_scale_w=tts_info["noise_scale_w"],
speed=tts_info["length_scale"])
speed=tts_info["speed"])
else:
audio,sr = tts.synthesize(text=sentence,
noise_scale=tts_info["noise_scale"],
noise_scale_w=tts_info["noise_scale_w"],
speed=tts_info["length_scale"],
speed=tts_info["speed"],
target_se=target_se)
text_response = {"type": "llm_text", "code": 200, "msg": sentence}
await ws.send_bytes(audio) #返回音频二进制流数据
await ws.send_text(json.dumps(text_response, ensure_ascii=False)) #返回文本数据
logger.debug(f"llm返回结果: {sentence}")
logger.debug(f"websocket返回: {sentence}")
if is_end:
logger.debug(f"llm返回结果: {llm_response}")
await ws.send_text(json.dumps({"type": "end", "code": 200, "msg": ""}, ensure_ascii=False))