EngineX-Iluvatar/enginex_bi_series-sherpa-onnx
Archived
16
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Add C API for audio tagging (#754)

Browse Source
This commit is contained in:
Fangjun Kuang
2024-04-11 14:18:43 +08:00
committed by GitHub
parent 34d70a259f
commit f204e62b44
9 changed files with 289 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
c-api-examples/CMakeLists.txt
View File

@@ -18,6 +18,9 @@ target_link_libraries(speaker-identification-c-api sherpa-onnx-c-api)
add_executable(streaming-hlg-decode-file-c-api streaming-hlg-decode-file-c-api.c)
target_link_libraries(streaming-hlg-decode-file-c-api sherpa-onnx-c-api)
add_executable(audio-tagging-c-api audio-tagging-c-api.c)
target_link_libraries(audio-tagging-c-api sherpa-onnx-c-api)
if(SHERPA_ONNX_HAS_ALSA)
add_subdirectory(./asr-microphone-example)
elseif((UNIX AND NOT APPLE) OR LINUX)

79
c-api-examples/audio-tagging-c-api.c Normal file
View File

@@ -0,0 +1,79 @@
// c-api-examples/audio-tagging-c-api.c
//
// Copyright (c) 2024 Xiaomi Corporation
// We assume you have pre-downloaded the model files for testing
// from https://github.com/k2-fsa/sherpa-onnx/releases/tag/audio-tagging-models
//
// An example is given below:
//
// clang-format off
//
// wget https://github.com/k2-fsa/sherpa-onnx/releases/download/audio-tagging-models/sherpa-onnx-zipformer-audio-tagging-2024-04-09.tar.bz2
// tar xvf sherpa-onnx-zipformer-audio-tagging-2024-04-09.tar.bz2
// rm sherpa-onnx-zipformer-audio-tagging-2024-04-09.tar.bz2
//
// clang-format on
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "sherpa-onnx/c-api/c-api.h"
int32_t main() {
SherpaOnnxAudioTaggingConfig config;
memset(&config, 0, sizeof(config));
config.model.zipformer.model =
"./sherpa-onnx-zipformer-audio-tagging-2024-04-09/model.int8.onnx";
config.model.num_threads = 1;
config.model.debug = 1;
config.model.provider = "cpu";
// clang-format off
config.labels = "./sherpa-onnx-zipformer-audio-tagging-2024-04-09/class_labels_indices.csv";
// clang-format on
const SherpaOnnxAudioTagging *tagger = SherpaOnnxCreateAudioTagging(&config);
if (!tagger) {
fprintf(stderr, "Failed to create audio tagger. Please check your config");
return -1;
}
// You can find more test waves from
// https://github.com/k2-fsa/sherpa-onnx/releases/download/audio-tagging-models/sherpa-onnx-zipformer-audio-tagging-2024-04-09.tar.bz2
const char *wav_filename =
"./sherpa-onnx-zipformer-audio-tagging-2024-04-09/test_wavs/1.wav";
const SherpaOnnxWave *wave = SherpaOnnxReadWave(wav_filename);
if (wave == NULL) {
fprintf(stderr, "Failed to read %s\n", wav_filename);
return -1;
}
const SherpaOnnxOfflineStream *stream =
SherpaOnnxAudioTaggingCreateOfflineStream(tagger);
AcceptWaveformOffline(stream, wave->sample_rate, wave->samples,
wave->num_samples);
int32_t top_k = 5;
const SherpaOnnxAudioEvent *const *results =
SherpaOnnxAudioTaggingCompute(tagger, stream, top_k);
fprintf(stderr, "--------------------------------------------------\n");
fprintf(stderr, "Index\t\tProbability\t\tEvent name\n");
fprintf(stderr, "--------------------------------------------------\n");
for (int32_t i = 0; i != top_k; ++i) {
fprintf(stderr, "%d\t\t%.3f\t\t\t%s\n", i, results[i]->prob,
results[i]->name);
}
fprintf(stderr, "--------------------------------------------------\n");
SherpaOnnxAudioTaggingFreeResults(results);
DestroyOfflineStream(stream);
SherpaOnnxFreeWave(wave);
SherpaOnnxDestroyAudioTagging(tagger);
return 0;
};