dearwolf/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2025-05-30 15:35:01 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 25

feat(llama.cpp): add support for audio input (#5466)

Browse source 
* feat(llama.cpp): add support for audio input

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* Adapt tests

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2025-05-26 16:06:03 +02:00 • committed by GitHub
parent 9650d490d4
commit 88de2ea01a
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 40 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

37
backend/cpp/llama/grpc-server.cpp
View file

@ -133,6 +133,15 @@ json parse_options(bool streaming, const backend::PredictOptions* predict)
});
}
// for each audio in the request, add the audio data
for (int i = 0; i < predict->audios_size(); i++) {
data["audio_data"].push_back(json
{
{"id", i},
{"data", predict->audios(i)},
});
}
data["stop"] = predict->stopprompts();
// data["n_probs"] = predict->nprobs();
//TODO: images,
@ -406,6 +415,16 @@ public:
}
}
const auto &audio_data = data.find("audio_data");
if (audio_data != data.end() && audio_data->is_array())
{
for (const auto &audio : *audio_data)
{
auto decoded_data = base64_decode(audio["data"].get<std::string>());
files.push_back(decoded_data);
}
}
// process files
mtmd::bitmaps bitmaps;
const bool has_mtmd = ctx_server.mctx != nullptr;
@ -416,10 +435,10 @@ public:
for (auto & file : files) {
mtmd::bitmap bmp(mtmd_helper_bitmap_init_from_buf(file.data(), file.size()));
if (!bmp.ptr) {
throw std::runtime_error("Failed to load image");
throw std::runtime_error("Failed to load image/audio");
}
// calculate bitmap hash (for KV caching)
std::string hash = fnv_hash(bmp.data(), bmp.nx()*bmp.ny()*3);
std::string hash = fnv_hash(bmp.data(), bmp.n_bytes());
bmp.set_id(hash.c_str());
bitmaps.entries.push_back(std::move(bmp));
}
@ -588,6 +607,16 @@ public:
}
}
const auto &audio_data = data.find("audio_data");
if (audio_data != data.end() && audio_data->is_array())
{
for (const auto &audio : *audio_data)
{
auto decoded_data = base64_decode(audio["data"].get<std::string>());
files.push_back(decoded_data);
}
}
// process files
mtmd::bitmaps bitmaps;
const bool has_mtmd = ctx_server.mctx != nullptr;
@ -598,10 +627,10 @@ public:
for (auto & file : files) {
mtmd::bitmap bmp(mtmd_helper_bitmap_init_from_buf(file.data(), file.size()));
if (!bmp.ptr) {
throw std::runtime_error("Failed to load image");
throw std::runtime_error("Failed to load image/audio");
}
// calculate bitmap hash (for KV caching)
std::string hash = fnv_hash(bmp.data(), bmp.nx()*bmp.ny()*3);
std::string hash = fnv_hash(bmp.data(), bmp.n_bytes());
bmp.set_id(hash.c_str());
bitmaps.entries.push_back(std::move(bmp));
}