From 3f260b924cb6b56d889e78fd6612c3214c76ea18 Mon Sep 17 00:00:00 2001
From: zhousha <736730048@qq.com>
Date: Thu, 14 Aug 2025 09:56:11 +0800
Subject: [PATCH] Add bi_v100-sensevoice and bi_v100-whisper as subdirectories

---
 .DS_Store                                     | Bin 6148 -> 6148 bytes
 README.md                                     |  84 ++++++++++++++++--
 .../Dockerfile_sensevoice                     |   0
 bi_v100-sensevoice/README.md                  |   5 ++
 .../constraints_sensevoice.txt                |   0
 .../launch_sensevoice.sh                      |   0
 .../requirements_sensevoice.txt               |   0
 .../server_sensevoice.py                      |   0
 bi_v100-whisper/.DS_Store                     | Bin 0 -> 6148 bytes
 .../Dockerfile_whisper                        |   0
 .../asr_server_whisper.py                     |   0
 .../launch_service                            |   0
 .../whisper-tiny}/.msc                        | Bin
 .../whisper-tiny}/.mv                         |   0
 14 files changed, 82 insertions(+), 7 deletions(-)
 rename Dockerfile_sensevoice => bi_v100-sensevoice/Dockerfile_sensevoice (100%)
 create mode 100644 bi_v100-sensevoice/README.md
 rename constraints_sensevoice.txt => bi_v100-sensevoice/constraints_sensevoice.txt (100%)
 rename launch_sensevoice.sh => bi_v100-sensevoice/launch_sensevoice.sh (100%)
 rename requirements_sensevoice.txt => bi_v100-sensevoice/requirements_sensevoice.txt (100%)
 rename server_sensevoice.py => bi_v100-sensevoice/server_sensevoice.py (100%)
 create mode 100644 bi_v100-whisper/.DS_Store
 rename Dockerfile_whisper => bi_v100-whisper/Dockerfile_whisper (100%)
 rename asr_server_whisper.py => bi_v100-whisper/asr_server_whisper.py (100%)
 rename launch_service => bi_v100-whisper/launch_service (100%)
 rename {whisper-tiny => bi_v100-whisper/whisper-tiny}/.msc (100%)
 rename {whisper-tiny => bi_v100-whisper/whisper-tiny}/.mv (100%)

diff --git a/.DS_Store b/.DS_Store
index 4dbf4df4ec57de0ac65234274a17244e15053ef3..debd41f52aa64ee4ac004fa43b56b99b7a756da7 100644
GIT binary patch
delta 161
zcmZoMXfc=|&e%S&P;8=}q9_vs0|O%ig8&1AA%g)B>M|5Fq)yzZKG{HoB|?ZHi6N6A
zo}mm`k|7VI1SpmdluKqvO({;!Ny^X90UE@>kD;NQAp@wan4tiur3j{U<3<Jc&FmZk
a9E|N7H-2ZH%rBzL$gtT_<S_HZ1{MH*&?K(_

delta 102
zcmZoMXfc=|&e%4wP;8=}q9`K+0|O8XFfbG_q%ssS=rWWrWKR63E(?<2VJK(FV8~=B
rMwVpAW2oGCUxR%Tqxoib4gn6vwv8XZGf(Ch(PahcIk4GQ<S;V;bX^z5

diff --git a/README.md b/README.md
index 3ac1ebc..c985a00 100644
--- a/README.md
+++ b/README.md
@@ -1,13 +1,83 @@
-# tiangai100-sensevoice-funasr
-# tiangai100-whisper
+# tiangai100-sensevoice-funasr & tiangai100-whisper
 
-【语音识别】
+## 天数智芯 天垓100 语音识别
+该模型测试框架在天垓100加速卡上，适配 sensevoice 和 whisper 大类下的模型，将语音信号转换为文本。
 
-funasr可正常照nvidia方式使用。
+SenseVoice 是由 FunAudioLLM 开发的一款开源音频处理模型，是阿里巴巴通义实验室推出的语音技术框架的核心模型之一，为高效、准确的多语言语音识别和音频分析工具。
+该 sensevoice 测试框架支持 Sensevoice-Small 和 Sensevoice-Large 模型。
 
-sensevoice测试框架支持Sensevoice-Small和Sensevoice-Large模型
+whisper 模型是 OpenAI 发布的一个通用语音识别模型，具有多语言支持、多任务学习等特点，采用编码器 - 解码器的 Transformer 结构。
+该whisper测试框架支持tiny、base、small、medium、large、large-v2、large-v3等。
 
-【天垓100上模型测试情况】
-| 模型大类   | 模型名称       | 天垓100运行时间/s | Nvidia A100运行时间/s |
+## sensevoice 模型测试服务原理
+通过 AutoModel （funasr 库提供的一个自动模型加载工具）加载 SenseVoice 模型，然后用 model.generate(...) 调用模型进行语音识别，其中 generate 是模型封装好的接口，内部实现了完整的语音识别流程。
+
+## whisper 模型测试服务原理
+pipeline 为 Hugging Face transformers 库提供的高层 API ，用于快速构建特定任务的处理管道。pipeline 内部封装了语音预处理（如特征提取）、模型推理、结果后处理（如文本解码）等步骤，简化了语音识别的调用流程。代码中调用 pipeline 自动完成从音频数据到文本的转换。
+
+## 如何使用 sensevoice 和 whisper 测试框架
+代码实现了一个接收音频数据并返回识别文本的语音识别 HTTP 服务，并基于 corex:3.2.1 基础镜像，将该 HTTP 服务重新打包成 docker 镜像，通过 k8s 集群sut容器去请求这个 HTTP 服务。
+
+## sensevoice 和 whisper 语音识别模型测试服务请求示例
+```python
+import requests  
+
+# 服务健康检查
+response = requests.get(f"{sut_url}/health")
+print(response.status_code, response.text)
+
+# 服务地址（根据实际部署情况修改IP和端口,端口为80）
+url = f"{sut_url}/recognition"
+#url = "http://localhost:80/recognition"
+
+# 读取音频文件（需是服务支持的格式，如wav等）
+with open("audio.wav", "rb") as f:
+    audio_data = f.read()
+
+# 请求参数（可选，指定语言，如中文为'zh'）
+params = {
+    "language": "zh"  # 支持的语言代码需参考whisper模型的支持列表
+}
+
+# 发送POST请求
+response = requests.post(
+    url,
+    data=audio_data,  # 音频数据放在请求体中
+    params=params     # 参数放在URL查询字符串中
+)
+# 请求体（data 参数）需要传入二进制音频数据
+# 语言参数（language）是可选的，默认值为 en（英语）
+# 支持的语言代码可参考 whisper 模型的官方文档（如中文是 zh，日语是 ja 等）
+# 确保音频文件格式被模型支持（whisper 通常支持 wav、mp3 等常见格式）
+
+# 解析响应结果
+if response.status_code == 200:
+    result = response.json()
+    print("识别状态:", result["RecognitionStatus"])
+    print("识别结果:", result["DisplayText"])
+else:
+    print("请求失败，状态码:", response.status_code)
+
+```
+成功时返回 JSON 格式的响应
+```json
+{
+    "RecognitionStatus": "Success",
+    "DisplayText": "识别到的文本内容"
+}
+```
+
+## 天垓100 SenseVoiceSmall 模型测试结果
+测试方式为在 天垓100 加速卡和 Nvidia A100 上对同一段音频（约10s）进行语音识别任务，获取运行时间
+| 模型类型   | 模型名称       | 天垓100运行时间/s | Nvidia A100运行时间/s |
 |------------|----------------|-------------------|-----------------------|
 | sensevoice | SenseVoiceSmall | 0.2               | 0.06                  |
+
+
+## 天垓100 whisper-tiny 模型测试结果
+测试方式为在 天垓100 加速卡和 CPU 上对同一段长音频（约9120s）进行语音识别任务，获取运行时间，1-cer指标
+| 模型类型 | 模型名称    | 天垓 A100对比CPU加速比 | 天垓 A100 1-cer | CPU 1-cer |
+|----------|-------------|------------------------|-----------------|-----------|
+| whisper  | whisper-tiny | 3.24x                  | 0.792           | 0.794     |
+
+
diff --git a/Dockerfile_sensevoice b/bi_v100-sensevoice/Dockerfile_sensevoice
similarity index 100%
rename from Dockerfile_sensevoice
rename to bi_v100-sensevoice/Dockerfile_sensevoice
diff --git a/bi_v100-sensevoice/README.md b/bi_v100-sensevoice/README.md
new file mode 100644
index 0000000..a4ef1d3
--- /dev/null
+++ b/bi_v100-sensevoice/README.md
@@ -0,0 +1,5 @@
+# tiangai100-sensevoice-funasr
+
+【语音识别】
+
+funasr可正常照nvidia方式使用。
diff --git a/constraints_sensevoice.txt b/bi_v100-sensevoice/constraints_sensevoice.txt
similarity index 100%
rename from constraints_sensevoice.txt
rename to bi_v100-sensevoice/constraints_sensevoice.txt
diff --git a/launch_sensevoice.sh b/bi_v100-sensevoice/launch_sensevoice.sh
similarity index 100%
rename from launch_sensevoice.sh
rename to bi_v100-sensevoice/launch_sensevoice.sh
diff --git a/requirements_sensevoice.txt b/bi_v100-sensevoice/requirements_sensevoice.txt
similarity index 100%
rename from requirements_sensevoice.txt
rename to bi_v100-sensevoice/requirements_sensevoice.txt
diff --git a/server_sensevoice.py b/bi_v100-sensevoice/server_sensevoice.py
similarity index 100%
rename from server_sensevoice.py
rename to bi_v100-sensevoice/server_sensevoice.py
diff --git a/bi_v100-whisper/.DS_Store b/bi_v100-whisper/.DS_Store
new file mode 100644
index 0000000000000000000000000000000000000000..4dbf4df4ec57de0ac65234274a17244e15053ef3
GIT binary patch
literal 6148
zcmeHK%Sr=55Ukc50_KpT$N2#d{=pEEFYpIMBPawTMiFm&^1J*rs~-r<#*5%Zs-e4P
zdb@U*I&5zPu+{7K5?BJ5(;e~QVQT)|ePSn-F(RGsc)}f?ae)^+OsbD3oV&mq?#Vce
z_>Xws9&UFx`~B`XRaq$@1*Cu!kOETR7YcaqrOnS06{Ua_kOJQd`1hgF9ed%B7@rP?
z7y*bgro*_7S%TO+LF|P?A~Q5gDlw^6BZeiN`Brtka7au#tcDM(ldUEci_`i37Ui&>
zs3--bz*vFP+%CQUZ|FbF|6`JNQa}p)D+O$}zFMz%rRuGd%XzPD^lQ4;eA3;x4hloG
lV`8*pZoD0zM^V-_U-Nk{91?@he9(#d8E{=>QsA!@_yRcE7BBz+

literal 0
HcmV?d00001

diff --git a/Dockerfile_whisper b/bi_v100-whisper/Dockerfile_whisper
similarity index 100%
rename from Dockerfile_whisper
rename to bi_v100-whisper/Dockerfile_whisper
diff --git a/asr_server_whisper.py b/bi_v100-whisper/asr_server_whisper.py
similarity index 100%
rename from asr_server_whisper.py
rename to bi_v100-whisper/asr_server_whisper.py
diff --git a/launch_service b/bi_v100-whisper/launch_service
similarity index 100%
rename from launch_service
rename to bi_v100-whisper/launch_service
diff --git a/whisper-tiny/.msc b/bi_v100-whisper/whisper-tiny/.msc
similarity index 100%
rename from whisper-tiny/.msc
rename to bi_v100-whisper/whisper-tiny/.msc
diff --git a/whisper-tiny/.mv b/bi_v100-whisper/whisper-tiny/.mv
similarity index 100%
rename from whisper-tiny/.mv
rename to bi_v100-whisper/whisper-tiny/.mv