From 43659b9882dfde6aa11ece88189a26a86479cc04 Mon Sep 17 00:00:00 2001
From: KP <109694228@qq.com>
Date: Tue, 19 Apr 2022 18:31:05 +0800
Subject: [PATCH] Add KWS example.

---
 examples/hey_snips/README.md      | 24 +++++-------------------
 examples/hey_snips/RESULTS.md     |  8 --------
 examples/hey_snips/kws0/README.md | 22 ++++++++++++++++++++++
 3 files changed, 27 insertions(+), 27 deletions(-)
 delete mode 100644 examples/hey_snips/RESULTS.md
 create mode 100644 examples/hey_snips/kws0/README.md

diff --git a/examples/hey_snips/README.md b/examples/hey_snips/README.md
index be8d142b..ba263906 100644
--- a/examples/hey_snips/README.md
+++ b/examples/hey_snips/README.md
@@ -1,22 +1,8 @@
-# MDTC Keyword Spotting with HeySnips Dataset
 
-## Dataset
+## Metrics
 
-Before running scripts, you **MUST** follow this instruction to download the dataset: https://github.com/sonos/keyword-spotting-research-datasets
+We mesure FRRs with fixing false alarms in one hour:
 
-After you download and decompress the dataset archive, you should **REPLACE** the value of `data_dir` in `conf/*.yaml` to complete dataset config.
-
-## Get Started
-
-In this section, we will train the [MDTC](https://arxiv.org/pdf/2102.13552.pdf) model and evaluate on "Hey Snips" dataset.
-
-```sh
-CUDA_VISIBLE_DEVICES=0,1 ./run.sh conf/mdtc.yaml
-```
-
-This script contains training and scoring steps. You can just set the `CUDA_VISIBLE_DEVICES` environment var to run on single gpu or multi-gpus.
-
-The vars `stage` and `stop_stage` in `./run.sh` controls the running steps:
-- stage 1: Training from scratch.
-- stage 2: Evaluating model on test dataset and computing detection error tradeoff(DET) of all trigger thresholds.
-- stage 3: Plotting the DET cruve for visualizaiton.
+|Model|False Alarm| False Reject Rate|
+|--|--|--|
+|MDTC| 1| 0.003559 |
diff --git a/examples/hey_snips/RESULTS.md b/examples/hey_snips/RESULTS.md
deleted file mode 100644
index ba263906..00000000
--- a/examples/hey_snips/RESULTS.md
+++ /dev/null
@@ -1,8 +0,0 @@
-
-## Metrics
-
-We mesure FRRs with fixing false alarms in one hour:
-
-|Model|False Alarm| False Reject Rate|
-|--|--|--|
-|MDTC| 1| 0.003559 |
diff --git a/examples/hey_snips/kws0/README.md b/examples/hey_snips/kws0/README.md
new file mode 100644
index 00000000..be8d142b
--- /dev/null
+++ b/examples/hey_snips/kws0/README.md
@@ -0,0 +1,22 @@
+# MDTC Keyword Spotting with HeySnips Dataset
+
+## Dataset
+
+Before running scripts, you **MUST** follow this instruction to download the dataset: https://github.com/sonos/keyword-spotting-research-datasets
+
+After you download and decompress the dataset archive, you should **REPLACE** the value of `data_dir` in `conf/*.yaml` to complete dataset config.
+
+## Get Started
+
+In this section, we will train the [MDTC](https://arxiv.org/pdf/2102.13552.pdf) model and evaluate on "Hey Snips" dataset.
+
+```sh
+CUDA_VISIBLE_DEVICES=0,1 ./run.sh conf/mdtc.yaml
+```
+
+This script contains training and scoring steps. You can just set the `CUDA_VISIBLE_DEVICES` environment var to run on single gpu or multi-gpus.
+
+The vars `stage` and `stop_stage` in `./run.sh` controls the running steps:
+- stage 1: Training from scratch.
+- stage 2: Evaluating model on test dataset and computing detection error tradeoff(DET) of all trigger thresholds.
+- stage 3: Plotting the DET cruve for visualizaiton.
-- 
GitLab