From 3cd3473bf7a3b41484baa86d9092248d78e7af39 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期五, 21 四月 2023 17:17:37 +0800
Subject: [PATCH] docs
---
egs_modelscope/speaker_verification/speech_xvector_sv-en-us-callhome-8k-spk6135-pytorch/infer.py | 10 +++++-----
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/egs_modelscope/speaker_verification/speech_xvector_sv-en-us-callhome-8k-spk6135-pytorch/infer.py b/egs_modelscope/speaker_verification/speech_xvector_sv-en-us-callhome-8k-spk6135-pytorch/infer.py
index d3975ae..1fd9dc6 100644
--- a/egs_modelscope/speaker_verification/speech_xvector_sv-en-us-callhome-8k-spk6135-pytorch/infer.py
+++ b/egs_modelscope/speaker_verification/speech_xvector_sv-en-us-callhome-8k-spk6135-pytorch/infer.py
@@ -11,24 +11,24 @@
# extract speaker embedding
# for url use "spk_embedding" as key
rec_result = inference_sv_pipline(
- audio_in='https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/sv_example_enroll.wav')
+ audio_in='https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_data/sv_example_enroll.wav')
enroll = rec_result["spk_embedding"]
# for local file use "spk_embedding" as key
- rec_result = inference_sv_pipline(audio_in='sv_example_same.wav')["test1"]
+ rec_result = inference_sv_pipline(audio_in='example/sv_example_same.wav')
same = rec_result["spk_embedding"]
import soundfile
- wav = soundfile.read('sv_example_enroll.wav')[0]
+ wav = soundfile.read('example/sv_example_enroll.wav')[0]
# for raw inputs use "spk_embedding" as key
spk_embedding = inference_sv_pipline(audio_in=wav)["spk_embedding"]
rec_result = inference_sv_pipline(
- audio_in='https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/sv_example_different.wav')
+ audio_in='https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_data/sv_example_different.wav')
different = rec_result["spk_embedding"]
# calculate cosine similarity for same speaker
- sv_threshold = 0.9465
+ sv_threshold = 0.80
same_cos = np.sum(enroll * same) / (np.linalg.norm(enroll) * np.linalg.norm(same))
same_cos = max(same_cos - sv_threshold, 0.0) / (1.0 - sv_threshold) * 100.0
print("Similarity:", same_cos)
--
Gitblit v1.9.1