rename to WhisperService

Author: litongjava

CMakeLists.txt

@@ -34,7 +34,7 @@ target_link_libraries(sdl_version ${SDL2_LIBRARIES})
 add_executable(simplest simplest.cpp common.cpp)
 target_link_libraries(simplest whisper)
 
-add_executable(stream_components stream_components.cpp stream_components_audio.cpp stream_components_output.cpp stream_components_service.cpp)
+add_executable(stream_components stream_local.cpp stream_components_audio.cpp stream_components_output.cpp stream_components_service.cpp)
 target_link_libraries(stream_components ${SDL2_LIBRARIES})
 
 add_executable(server server.cpp common.cpp httplib.h json.hpp inference_handler.cpp whisper_params.cpp)

README.md

@@ -1,4 +1,4 @@
-# whisper-cpp server
+# whisper-cpp service
 ## simplest
 ```shell
 cmake-build-debug/simplest -m models/ggml-base.en.bin samples/jfk.wav
@@ -11,12 +11,12 @@ options:
 -m FNAME,  --model FNAME       [models/ggml-base.en.bin] model path                                                                                                                                   
 -di,       --diarize           [false  ] stereo audio diarization
 ```
-## server
+## service
 
-Simple http server. WAV Files are passed to the inference model via http requests.
+Simple http service. WAV Files are passed to the inference model via http requests.
 
 ```
-./cmake-build-debug/server -m models/ggml-base.en.bin
+./cmake-build-debug/service -m models/ggml-base.en.bin
 ```
 
 ```shell
@@ -81,7 +81,7 @@ ggml_metal_add_buffer: allocated 'backend         ' buffer, size =     4.05 MB,
 ggml_metal_add_buffer: allocated 'backend         ' buffer, size =     3.08 MB, (  203.67 / 11453.25)
 ggml_metal_add_buffer: allocated 'backend         ' buffer, size =    94.78 MB, (  298.45 / 11453.25)
 
-whisper server listening at http://0.0.0.0:8080
+whisper service listening at http://0.0.0.0:8080
 
 Received request: jfk.wav
 Successfully loaded jfk.wav
@@ -95,9 +95,9 @@ Running whisper.cpp inference on jfk.wav
 [00:00:00.000 --> 00:00:11.000]   And so my fellow Americans, ask not what your country can do for you, ask what you can do for your country.
 ```
 ```
-./server -h
+./service -h
 
-usage: ./bin/server [options]
+usage: ./bin/service [options]
 
 options:
   -h,        --help              [default] show this help message and exit
@@ -128,8 +128,8 @@ options:
              --prompt PROMPT     [       ] initial prompt
   -m FNAME,  --model FNAME       [models/ggml-base.en.bin] model path
   -oved D,   --ov-e-device DNAME [CPU    ] the OpenVINO device used for encode inference
-  --host HOST,                   [127.0.0.1] Hostname/ip-adress for the server
-  --port PORT,                   [8080   ] Port number for the server
+  --host HOST,                   [127.0.0.1] Hostname/ip-adress for the service
+  --port PORT,                   [8080   ] Port number for the service
 ```
 
 ## request examples

httplib.h

@@ -468,7 +468,7 @@ namespace httplib {
     std::string local_addr;
     int local_port = -1;
 
-    // for server
+    // for service
     std::string version;
     std::string target;
     Params params;
@@ -1895,7 +1895,7 @@ class SSLClient : public ClientImpl {
       case Error::Canceled: return "Connection handling canceled";
       case Error::SSLConnection: return "SSL connection failed";
       case Error::SSLLoadingCerts: return "SSL certificate loading failed";
-      case Error::SSLServerVerification: return "SSL server verification failed";
+      case Error::SSLServerVerification: return "SSL service verification failed";
       case Error::UnsupportedMultipartBoundaryChars:
         return "Unsupported HTTP multipart boundary characters";
       case Error::Compression: return "Compression failed";
@@ -5501,7 +5501,7 @@ namespace detail {
 
 } // namespace detail
 
-// HTTP server implementation
+// HTTP service implementation
 inline Server::Server()
   : new_task_queue(
   [] { return new ThreadPool(CPPHTTPLIB_THREAD_POOL_COUNT); }) {
@@ -6187,7 +6187,7 @@ inline bool Server::listen_internal() {
           detail::close_socket(svr_sock_);
           ret = false;
         } else {
-          ; // The server socket was closed by user.
+          ; // The service socket was closed by user.
         }
         break;
       }
@@ -8238,7 +8238,7 @@ static SSLInit sslinit_;
 
 } // namespace detail
 
-// SSL HTTP server implementation
+// SSL HTTP service implementation
 inline SSLServer::SSLServer(const char *cert_path, const char *private_key_path,
                             const char *client_ca_cert_file_path,
                             const char *client_ca_cert_dir_path,

inference_handler.cpp

@@ -7,7 +7,7 @@
 using json = nlohmann::json;
 
 struct whisper_print_user_data {
-  const whisper_params *params;
+  const whisper_local_stream_params *params;
 
   const std::vector<std::vector<float>> *pcmf32s;
   int progress_prev;
@@ -149,7 +149,7 @@ void whisper_print_segment_callback(struct whisper_context *ctx, struct whisper_
 }
 
 std::string
-output_str(struct whisper_context *ctx, const whisper_params &params, std::vector<std::vector<float>> pcmf32s) {
+output_str(struct whisper_context *ctx, const whisper_local_stream_params &params, std::vector<std::vector<float>> pcmf32s) {
   std::stringstream result;
   const int n_segments = whisper_full_n_segments(ctx);
   for (int i = 0; i < n_segments; ++i) {
@@ -178,7 +178,7 @@ void whisper_print_progress_callback(struct whisper_context * /*ctx*/, struct wh
   }
 }
 
-void getReqParameters(const Request &req, whisper_params &params) {
+void getReqParameters(const Request &req, whisper_local_stream_params &params) {
 // user model configu.has_fileion
   if (req.has_file("offset-t")) {
     params.offset_t_ms = std::stoi(req.get_file_value("offset-t").content);
@@ -204,9 +204,9 @@ void getReqParameters(const Request &req, whisper_params &params) {
 }
 
 
-void getReqParameters(const Request &request, whisper_params &params);
+void getReqParameters(const Request &request, whisper_local_stream_params &params);
 
-void handleInference(const Request &req, Response &res, std::mutex &whisper_mutex, whisper_params &params,
+void handleInference(const Request &req, Response &res, std::mutex &whisper_mutex, whisper_local_stream_params &params,
                      whisper_context *ctx, char *arg_audio_file) {
 // aquire whisper model mutex lock
   whisper_mutex.lock();

inference_handler.h

@@ -5,5 +5,5 @@
 
 using namespace httplib;
 
-void handleInference(const Request &req, Response &res, std::mutex &whisper_mutex, whisper_params &params,
+void handleInference(const Request &req, Response &res, std::mutex &whisper_mutex, whisper_local_stream_params &params,
                      whisper_context *ctx, char *arg_audio_file);

server.cpp

@@ -19,7 +19,7 @@ bool is_file_exist(const char *fileName) {
 }
 
 int main(int argc, char **argv) {
-  whisper_params params;
+  whisper_local_stream_params params;
   server_params sparams;
 
   std::mutex whisper_mutex;
@@ -138,7 +138,7 @@ int main(int argc, char **argv) {
   svr.set_write_timeout(sparams.write_timeout);
 
   if (!svr.bind_to_port(sparams.hostname, sparams.port)) {
-    fprintf(stderr, "\ncouldn't bind to server socket: hostname=%s port=%d\n\n",
+    fprintf(stderr, "\ncouldn't bind to service socket: hostname=%s port=%d\n\n",
             sparams.hostname.c_str(), sparams.port);
     return 1;
   }
@@ -147,7 +147,7 @@ int main(int argc, char **argv) {
   svr.set_base_dir(sparams.public_path);
 
   // to make it ctrl+clickable:
-  printf("\nwhisper server listening at http://%s:%d\n\n", sparams.hostname.c_str(), sparams.port);
+  printf("\nwhisper service listening at http://%s:%d\n\n", sparams.hostname.c_str(), sparams.port);
 
   if (!svr.listen_after_bind()) {
     return 1;

simplest.cpp

@@ -52,7 +52,7 @@ void replace_all(std::string &s, const std::string &search, const std::string &r
 }
 
 // command-line parameters
-struct whisper_params {
+struct whisper_local_stream_params {
   int32_t n_threads = std::min(4, (int32_t) std::thread::hardware_concurrency());
   int32_t n_processors = 1;
   int32_t offset_t_ms = 0;
@@ -97,9 +97,9 @@ struct whisper_params {
   std::vector<std::string> fname_out = {};
 };
 
-void whisper_print_usage(int argc, char **argv, const whisper_params &params);
+void whisper_print_usage(int argc, char **argv, const whisper_local_stream_params &params);
 
-bool whisper_params_parse(int argc, char **argv, whisper_params &params) {
+bool whisper_params_parse(int argc, char **argv, whisper_local_stream_params &params) {
   for (int i = 1; i < argc; i++) {
     std::string arg = argv[i];
 
@@ -128,7 +128,7 @@ bool whisper_params_parse(int argc, char **argv, whisper_params &params) {
   return true;
 }
 
-void whisper_print_usage(int /*argc*/, char **argv, const whisper_params &params) {
+void whisper_print_usage(int /*argc*/, char **argv, const whisper_local_stream_params &params) {
   fprintf(stderr, "\n");
   fprintf(stderr, "usage: %s [options] file0.wav file1.wav ...\n", argv[0]);
   fprintf(stderr, "\n");
@@ -141,7 +141,7 @@ void whisper_print_usage(int /*argc*/, char **argv, const whisper_params &params
 }
 
 struct whisper_print_user_data {
-  const whisper_params *params;
+  const whisper_local_stream_params *params;
 
   const std::vector<std::vector<float>> *pcmf32s;
   int progress_prev;
@@ -298,7 +298,7 @@ char *escape_double_quotes_and_backslashes(const char *str) {
 
 int main(int argc, char **argv) {
   printf("start\n");
-  whisper_params params;
+  whisper_local_stream_params params;
 
   if (whisper_params_parse(argc, argv, params) == false) {
     whisper_print_usage(argc, argv, params);

stream_components_output.cpp

@@ -122,19 +122,19 @@ void WhisperEncoderJSON::doindent() {
 }
 
 
-// -- WhisperOutput --
+// -- WhisperStreamOutput --
 
-WhisperOutput::WhisperOutput(
+WhisperStreamOutput::WhisperStreamOutput(
   struct whisper_context *ctx,
-  const server_params &params) :
+  const service_params &params) :
   ctx(ctx),
   params(params) {
 
 }
 
-void WhisperOutput::encode_server(
+void WhisperStreamOutput::encode_server(
   WhisperEncoder &encoder,
-  const server_params &params,
+  const service_params &params,
   struct whisper_context *ctx) {
   encoder.reset();
 
@@ -167,7 +167,7 @@ void WhisperOutput::encode_server(
   encoder.end_obj(true);
 }
 
-void WhisperOutput::encode_transcription(WhisperEncoder &encoder) const {
+void WhisperStreamOutput::encode_transcription(WhisperEncoder &encoder) const {
   encoder.reset();
 
   encoder.start_obj(nullptr);
@@ -223,12 +223,12 @@ void WhisperOutput::encode_transcription(WhisperEncoder &encoder) const {
   encoder.end_obj(true);
 }
 
-void WhisperOutput::server_to_json(std::ostream &os, const server_params &params, struct whisper_context *ctx) {
+void WhisperStreamOutput::to_json(std::ostream &os, const service_params &params, struct whisper_context *ctx) {
   WhisperEncoderJSON encoder(os);
-  WhisperOutput::encode_server(encoder, params, ctx);
+  WhisperStreamOutput::encode_server(encoder, params, ctx);
 }
 
-void WhisperOutput::transcription_to_json(std::ostream &os) const {
+void WhisperStreamOutput::transcription_to_json(std::ostream &os) const {
   WhisperEncoderJSON encoder(os);
   encode_transcription(encoder);
 }

stream_components_output.h

@@ -7,8 +7,8 @@
 #include "stream_components_params.h"
 
 /**
- * Classes that support componentization of the server.
- * These classes support encoding the server state and the transcription
+ * Classes that support componentization of the service.
+ * These classes support encoding the service state and the transcription
  * into JSON (and in the future, into other formats such as msgpack,
  * flexbuffers, etc.)
  */
@@ -73,26 +73,26 @@ namespace stream_components {
     void doindent();
   };
 
-  class WhisperOutput {
+  class WhisperStreamOutput {
   public:
-    WhisperOutput(
+    WhisperStreamOutput(
       struct whisper_context *ctx,
-      const server_params &params);
+      const service_params &params);
 
-    static void encode_server(WhisperEncoder &encoder, const server_params &params, struct whisper_context *ctx);
+    static void encode_server(WhisperEncoder &encoder, const service_params &params, struct whisper_context *ctx);
 
     void encode_transcription(WhisperEncoder &encoder) const;
 
-    static void server_to_json(std::ostream &os, const server_params &params, struct whisper_context *ctx);
+    static void to_json(std::ostream &os, const service_params &params, struct whisper_context *ctx);
 
     void transcription_to_json(std::ostream &os) const;
 
   protected:
     struct whisper_context *ctx;
-    const server_params params;
+    const service_params params;
   };
 
-  using WhisperOutputPtr = std::shared_ptr<WhisperOutput>;
+  using WhisperOutputPtr = std::shared_ptr<WhisperStreamOutput>;
 
 } // namespace stream_components
 

stream_components_params.h

@@ -35,7 +35,7 @@ namespace stream_components {
     }
   };
 
-  struct server_params {
+  struct service_params {
     int32_t n_threads = std::min(4, (int32_t) std::thread::hardware_concurrency());
     bool speed_up = false;
     bool translate = false;