ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2024-12-24 06:46:37 +00:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

server : automatically convert audio on the server (#1539)

Browse Source 
* server : automatically convert audio on the server

* server : remove rebundant comments

* server : automatic conversion refactor

* server : update server readme

* server : remove unnecessary comments and tabs

* server : put back remove calling

* server : apply suggestions from code review

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* server : check ffmpeg before the server lunch

* server : fix indentation

* Apply suggestions from code review

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* server : fix function typo calling

* server : fix function typo calling

* server : add warning in readme

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
Ismatulla Mansurov 2023-11-27 02:28:34 -07:00 committed by GitHub
parent 447d49530c
commit 23c21e92eb
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 62 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
examples/server/README.md
View File

@ -43,8 +43,12 @@ options:
-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
--convert, [false ] Convert audio to WAV, requires ffmpeg on the server
```
> [!WARNING]
> **Do not run the server example with administrative privileges and ensure it's operated in a sandbox environment, especially since it involves risky operations like accepting user file uploads and using ffmpeg for format conversions. Always validate and sanitize inputs to guard against potential security threats.**
## request examples
**/inference**

59
examples/server/server.cpp
View File

@ -43,6 +43,8 @@ struct server_params
int32_t port = 8080;
int32_t read_timeout = 600;
int32_t write_timeout = 600;
bool ffmpeg_converter = false;
};
struct whisper_params {
@ -157,6 +159,7 @@ void whisper_print_usage(int /*argc*/, char ** argv, const whisper_params & para
fprintf(stderr, " --host HOST, [%-7s] Hostname/ip-adress for the server\n", sparams.hostname.c_str());
fprintf(stderr, " --port PORT, [%-7d] Port number for the server\n", sparams.port);
fprintf(stderr, " --public PATH, [%-7s] Path to the public folder\n", sparams.public_path.c_str());
fprintf(stderr, " --convert, [%-7s] Convert audio to WAV, requires ffmpeg on the server", sparams.ffmpeg_converter ? "true" : "false");
fprintf(stderr, "\n");
}
@ -203,6 +206,7 @@ bool whisper_params_parse(int argc, char ** argv, whisper_params & params, serve
else if ( arg == "--port") { sparams.port = std::stoi(argv[++i]); }
else if ( arg == "--host") { sparams.hostname = argv[++i]; }
else if ( arg == "--public") { sparams.public_path = argv[++i]; }
else if ( arg == "--convert") { sparams.ffmpeg_converter = true; }
else {
fprintf(stderr, "error: unknown argument: %s\n", arg.c_str());
whisper_print_usage(argc, argv, params, sparams);
@ -220,6 +224,45 @@ struct whisper_print_user_data {
int progress_prev;
};
void check_ffmpeg_availibility() {
int result = system("ffmpeg -version");
if (result == 0) {
std::cout << "ffmpeg is available." << std::endl;
} else {
// ffmpeg is not available
std::cout << "ffmpeg is not found. Please ensure that ffmpeg is installed ";
std::cout << "and that its executable is included in your system's PATH. ";
exit(0);
}
}
bool convert_to_wav(const std::string & temp_filename, std::string & error_resp) {
std::ostringstream cmd_stream;
std::string converted_filename_temp = temp_filename + "_temp.wav";
cmd_stream << "ffmpeg -i \"" << temp_filename << "\" -ar 16000 -ac 1 -c:a pcm_s16le \"" << converted_filename_temp << "\" 2>&1";
std::string cmd = cmd_stream.str();
int status = std::system(cmd.c_str());
if (status != 0) {
error_resp = "{\"error\":\"FFmpeg conversion failed.\"}";
return false;
}
// Remove the original file
if (remove(temp_filename.c_str()) != 0) {
error_resp = "{\"error\":\"Failed to remove the original file.\"}";
return false;
}
// Rename the temporary file to match the original filename
if (rename(converted_filename_temp.c_str(), temp_filename.c_str()) != 0) {
error_resp = "{\"error\":\"Failed to rename the temporary file.\"}";
return false;
}
return true;
}
std::string estimate_diarization_speaker(std::vector<std::vector<float>> pcmf32s, int64_t t0, int64_t t1, bool id_only = false) {
std::string speaker = "";
const int64_t n_samples = pcmf32s[0].size();
@ -407,6 +450,9 @@ int main(int argc, char ** argv) {
exit(0);
}
if (sparams.ffmpeg_converter) {
check_ffmpeg_availibility();
}
// whisper init
struct whisper_context_params cparams;
cparams.use_gpu = params.use_gpu;
@ -462,6 +508,18 @@ int main(int argc, char ** argv) {
temp_file << audio_file.content;
temp_file.close();
// if file is not wav, convert to wav
if (sparams.ffmpeg_converter) {
std::string error_resp = "{\"error\":\"Failed to execute ffmpeg command.\"}";
const bool is_converted = convert_to_wav(temp_filename, error_resp);
if (!is_converted) {
res.set_content(error_resp, "application/json");
whisper_mutex.unlock();
return;
}
}
// read wav content into pcmf32
if (!::read_wav(temp_filename, pcmf32, pcmf32s, params.diarize)) {
fprintf(stderr, "error: failed to read WAV file '%s'\n", temp_filename.c_str());
@ -509,7 +567,6 @@ int main(int argc, char ** argv) {
// run the inference
{
printf("Running whisper.cpp inference on %s\n", filename.c_str());
whisper_full_params wparams = whisper_full_default_params(WHISPER_SAMPLING_GREEDY);