whipser-asr-server

在线文档

https://apifox.com/apidoc/shared-98cc5675-f1a3-4250-a940-cfe060854ef4/api-121475073

接口

测试接口

测试接口无需上传文件,使用内部自带文件进行识别 访问地址:http://localhost/whispser/asr/test 返回数据

{
    "data": [
        {
            "end": 1088,
            "sentence": " And so, my fellow Americans, ask not what your country can do for you, ask what you can do for your country.",
            "start": 0
        }
    ],
    "ok": true
}

识别接口

识别接口需要上传文件音频文件 访问地址:http://localhost/whispser/asr/rec

POST /whispser/asr/test

Body Parameters

file: filedata
inputType: wav
outputType: default

Params

Name	Location	Type	Required	Description
body	body	object	no	none
» file	body	string(binary)	yes	上传的音频文件
» inputType	body	string	no	上传的音频格式wav和mp3
» outputType	body	string	no	返回的文本格式,支持default,irc,vtt,srt
» outputFormat	body	string	no	返回的数据格式,输出的格式,默认是json,如果需要字幕数据输出txt

Response Examples

成功

{
  "data": [
    {
      "end": 800,
      "sentence": " And so my fellow Americans, ask not what your country can do for you.",
      "start": 0
    },
    {
      "end": 1100,
      "sentence": " Ask what you can do for your country.",
      "start": 800
    }
  ],
  "ok": true
}

required

/usr/lib/x86_64-linux-gnu/libstdc++.so.6: version `GLIBCXX_3.4.21'

build

# Set java version
export JAVA_HOME=/usr/java/jdk-11.0.8
export PATH=$JAVA_HOME/bin:$PATH

#build jar
mvn clean install -DskipTests -Dgpg.skip -Pdevelopment
mvn clean package -DskipTests -Dgpg.skip -Pproduction -pl whisper-asr-server

run

java -jar whisper-asr-server/target/whisper-asr-server-1.0.1.jar

The default loaded model is /root/.cache/whisper/ggml-base.en.bin downlaod model form huggingface https://huggingface.co/ggerganov/whisper.cpp

specify the model name

java -jar whisper-asr-server/target/whisper-asr-server-1.0.1.jar --model.name=ggml-large.bin

convert file with ffmpeg

Note that the server currently runs only with 16-bit audio files, so make sure to convert your input before running the tool. For example, you can use ffmpeg like this:

ffmpeg -i input.mp3 -ar 16000 -ac 1 -c:a pcm_s16le output.wav

ffmpeg -i input.mp4 -ar 16000 -ac 1 -c:a pcm_s16le output.wav

构建Docker镜像

封装镜像

build

docker build -f docker/1.0.1 -t litongjava/whisper-asr-server:1.0.1 .

run

docker run -dit -p 8080:80 litongjava/whisper-asr-server:1.0.1

test

curl -v http://localhost:8080/whispser/asr/test