Hass/xiaozhi-esp32-server
1
0
Fork 0
mirror of https://github.com/xinnan-tech/xiaozhi-esp32-server.git synced 2026-07-22 07:03:53 +08:00
Code Issues Packages Projects Releases Wiki Activity

update：优化asr测速源：从已有的几个文件中测速度

Browse Source
This commit is contained in:
hrz
2025-05-20 11:14:20 +08:00
parent 84ac1b6686
commit 5360cbf19c
2 changed files with 32 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files
main/xiaozhi-server/config.yaml
+1 -1
View File
@@ -411,7 +411,7 @@ LLM:
base_url: https://host/api/v1
# 你可以在这里找到你的api_key
# https://cloud.tryfastgpt.ai/account/apikey
api_key: fastgpt-xxx
api_key: 你的fastgpt密钥
variables:
k: "v"
k2: "v2"
main/xiaozhi-server/performance_tester.py
+31 -25
View File
@@ -30,11 +30,13 @@ class AsyncPerformanceTester:
)
self.test_wav_list = []
self.wav_root = r".\config\assets\test_asr"
os.makedirs(self.wav_root, exist_ok=True)
for wav_name in os.listdir(self.wav_root):
with open(os.path.join(self.wav_root, wav_name), "rb") as f:
self.test_wav_list.append(f.read())
self.wav_root = r"config/assets"
for file_name in os.listdir(self.wav_root):
file_path = os.path.join(self.wav_root, file_name)
# 检查文件大小是否大于300KB
if os.path.getsize(file_path) > 300 * 1024: # 300KB = 300 * 1024 bytes
with open(file_path, "rb") as f:
self.test_wav_list.append(f.read())
self.results = {"llm": {}, "tts": {}, "stt": {}, "combinations": []}
@@ -73,9 +75,9 @@ class AsyncPerformanceTester:
token_fields = ["access_token", "api_key", "token"]
if any(
field in config
and any(x in config[field] for x in ["你的", "placeholder"])
for field in token_fields
field in config
and any(x in config[field] for x in ["你的", "placeholder"])
for field in token_fields
):
print(f"⏭️ TTS {tts_name} 未配置access_token/api_key，已跳过")
return {"name": tts_name, "type": "tts", "errors": 1}
@@ -185,7 +187,7 @@ class AsyncPerformanceTester:
return {"name": llm_name, "type": "llm", "errors": 1}
else:
if "api_key" in config and any(
x in config["api_key"] for x in ["你的", "placeholder", "sk-xxx"]
x in config["api_key"] for x in ["你的", "placeholder", "sk-xxx"]
):
print(f"🚫 跳过未配置的LLM: {llm_name}")
return {"name": llm_name, "type": "llm", "errors": 1}
@@ -257,7 +259,7 @@ class AsyncPerformanceTester:
async def process_response():
nonlocal first_token_received, first_token_time
for chunk in llm.response(
"perf_test", [{"role": "user", "content": sentence}]
"perf_test", [{"role": "user", "content": sentence}]
):
if not first_token_received and chunk.strip() != "":
first_token_time = time.time() - sentence_start
@@ -319,15 +321,15 @@ class AsyncPerformanceTester:
for stt in valid_stt:
# 计算相对性能分数（越小越好）
llm_score = (
self.results["llm"][llm]["avg_first_token"] / min_first_token
self.results["llm"][llm]["avg_first_token"] / min_first_token
)
tts_score = self.results["tts"][tts]["avg_time"] / min_tts_time
stt_score = self.results["stt"][stt]["avg_time"] / min_stt_time
# 计算稳定性分数（标准差/平均值，越小越稳定）
llm_stability = (
self.results["llm"][llm]["std_first_token"]
/ self.results["llm"][llm]["avg_first_token"]
self.results["llm"][llm]["std_first_token"]
/ self.results["llm"][llm]["avg_first_token"]
)
# 综合得分（考虑性能和稳定性）
@@ -335,7 +337,9 @@ class AsyncPerformanceTester:
llm_final_score = llm_score * 0.7 + llm_stability * 0.3
# 总分 = LLM得分(70%) + TTS得分(30%) + STT得分(30%)
total_score = llm_final_score * 0.7 + tts_score * 0.3 + stt_score * 0.3
total_score = (
llm_final_score * 0.7 + tts_score * 0.3 + stt_score * 0.3
)
self.results["combinations"].append(
{
@@ -484,12 +488,12 @@ class AsyncPerformanceTester:
# 检查配置有效性
if llm_name == "CozeLLM":
if any(x in config.get("bot_id", "") for x in ["你的"]) or any(
x in config.get("user_id", "") for x in ["你的"]
x in config.get("user_id", "") for x in ["你的"]
):
print(f"⏭️ LLM {llm_name} 未配置bot_id/user_id，已跳过")
continue
elif "api_key" in config and any(
x in config["api_key"] for x in ["你的", "placeholder", "sk-xxx"]
x in config["api_key"] for x in ["你的", "placeholder", "sk-xxx"]
):
print(f"⏭️ LLM {llm_name} 未配置api_key，已跳过")
continue
@@ -512,16 +516,18 @@ class AsyncPerformanceTester:
# 为每个句子创建独立任务
for sentence in self.test_sentences:
sentence = sentence.encode("utf-8").decode("utf-8")
all_tasks.append(self._test_single_sentence(llm_name, llm, sentence))
all_tasks.append(
self._test_single_sentence(llm_name, llm, sentence)
)
# TTS测试任务
if self.config.get("TTS") is not None:
for tts_name, config in self.config.get("TTS", {}).items():
token_fields = ["access_token", "api_key", "token"]
if any(
field in config
and any(x in config[field] for x in ["你的", "placeholder"])
for field in token_fields
field in config
and any(x in config[field] for x in ["你的", "placeholder"])
for field in token_fields
):
print(f"⏭️ TTS {tts_name} 未配置access_token/api_key，已跳过")
continue
@@ -534,9 +540,9 @@ class AsyncPerformanceTester:
for stt_name, config in self.config.get("ASR", {}).items():
token_fields = ["access_token", "api_key", "token"]
if any(
field in config
and any(x in config[field] for x in ["你的", "placeholder"])
for field in token_fields
field in config
and any(x in config[field] for x in ["你的", "placeholder"])
for field in token_fields
):
print(f"⏭️ ASR {stt_name} 未配置access_token/api_key，已跳过")
continue
@@ -587,9 +593,9 @@ class AsyncPerformanceTester:
"name": llm_name,
"type": "llm",
"avg_response": sum(data["response_times"])
/ len(data["response_times"]),
/ len(data["response_times"]),
"avg_first_token": sum(data["first_token_times"])
/ len(data["first_token_times"]),
/ len(data["first_token_times"]),
"std_first_token": (
statistics.stdev(data["first_token_times"])
if len(data["first_token_times"]) > 1