search

TrungThanhTran / VGGVoxRealTime

Thanks VGGVox Oxford Model for the Speaker Recognition model generation. We would like to port to Python for realtime running
6 stars 4 forks source link

model weights? #1

Open ghost opened 6 years ago

ghost commented 6 years ago

where can i download the model wights?

def test(): TEST_WAV1 = "data/wav/prdcv-script/4s_Linh0A_0.wav" TEST_WAV2 = "data/wav/prdcv-script/10s_Linh0A_0.wav" model = vggvox_model() model.load_weights("data/model_weights/model_0.h5") buckets = build_buckets(c.MAX_SEC_TEST,c.BUCKET_STEP_SEC) spec1 = read_and_process_audio(TEST_WAV1,buckets) emb1 = model.predict(spec1.reshape(1,spec1.shape,1)) spec2 = read_and_process_audio(TEST_WAV2,buckets) emb2 = model.predict(spec2.reshape(1,spec2.shape,1)) dist = np.linalg.norm(emb1-emb2) print(dist)

SwapnilBorse123 commented 5 years ago

Can you please add the data folder with files to the existing directory structure?