Support audio tagging using zipformer (#747)

This commit is contained in:
Fangjun Kuang
2024-04-10 14:47:06 +08:00
committed by GitHub
parent c9ae7595d5
commit f20291cadc
30 changed files with 927 additions and 11 deletions

32
.github/scripts/test-audio-tagging.sh vendored Executable file
View File

@@ -0,0 +1,32 @@
#!/usr/bin/env bash
set -ex
log() {
# This function is from espnet
local fname=${BASH_SOURCE[1]##*/}
echo -e "$(date '+%Y-%m-%d %H:%M:%S') (${fname}:${BASH_LINENO[0]}:${FUNCNAME[1]}) $*"
}
echo "EXE is $EXE"
echo "PATH: $PATH"
which $EXE
log "------------------------------------------------------------"
log "Run zipformer for audio tagging "
log "------------------------------------------------------------"
curl -SL -O 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
repo=sherpa-onnx-zipformer-audio-tagging-2024-04-09
ls -lh $repo
for w in 1.wav 2.wav 3.wav 4.wav; do
$EXE \
--zipformer-model=$repo/model.onnx \
--labels=$repo/class_labels_indices.csv \
$repo/test_wavs/$w
done
rm -rf $repo

View File

@@ -15,6 +15,7 @@ on:
- '.github/scripts/test-offline-ctc.sh'
- '.github/scripts/test-online-ctc.sh'
- '.github/scripts/test-offline-tts.sh'
- '.github/scripts/test-audio-tagging.sh'
- 'CMakeLists.txt'
- 'cmake/**'
- 'sherpa-onnx/csrc/*'
@@ -32,6 +33,7 @@ on:
- '.github/scripts/test-offline-ctc.sh'
- '.github/scripts/test-online-ctc.sh'
- '.github/scripts/test-offline-tts.sh'
- '.github/scripts/test-audio-tagging.sh'
- 'CMakeLists.txt'
- 'cmake/**'
- 'sherpa-onnx/csrc/*'
@@ -124,6 +126,14 @@ jobs:
name: release-${{ matrix.build_type }}-with-shared-lib-${{ matrix.shared_lib }}-with-tts-${{ matrix.with_tts }}
path: build/bin/*
- name: Test Audio tagging
shell: bash
run: |
export PATH=$PWD/build/bin:$PATH
export EXE=sherpa-onnx-offline-audio-tagging
.github/scripts/test-audio-tagging.sh
- name: Test online CTC
shell: bash
run: |

View File

@@ -15,6 +15,7 @@ on:
- '.github/scripts/test-offline-ctc.sh'
- '.github/scripts/test-offline-tts.sh'
- '.github/scripts/test-online-ctc.sh'
- '.github/scripts/test-audio-tagging.sh'
- 'CMakeLists.txt'
- 'cmake/**'
- 'sherpa-onnx/csrc/*'
@@ -31,6 +32,7 @@ on:
- '.github/scripts/test-offline-ctc.sh'
- '.github/scripts/test-offline-tts.sh'
- '.github/scripts/test-online-ctc.sh'
- '.github/scripts/test-audio-tagging.sh'
- 'CMakeLists.txt'
- 'cmake/**'
- 'sherpa-onnx/csrc/*'
@@ -103,6 +105,14 @@ jobs:
otool -L build/bin/sherpa-onnx
otool -l build/bin/sherpa-onnx
- name: Test Audio tagging
shell: bash
run: |
export PATH=$PWD/build/bin:$PATH
export EXE=sherpa-onnx-offline-audio-tagging
.github/scripts/test-audio-tagging.sh
- name: Test C API
shell: bash
run: |

View File

@@ -14,6 +14,7 @@ on:
- '.github/scripts/test-offline-ctc.sh'
- '.github/scripts/test-online-ctc.sh'
- '.github/scripts/test-offline-tts.sh'
- '.github/scripts/test-audio-tagging.sh'
- 'CMakeLists.txt'
- 'cmake/**'
- 'sherpa-onnx/csrc/*'
@@ -28,6 +29,7 @@ on:
- '.github/scripts/test-offline-ctc.sh'
- '.github/scripts/test-online-ctc.sh'
- '.github/scripts/test-offline-tts.sh'
- '.github/scripts/test-audio-tagging.sh'
- 'CMakeLists.txt'
- 'cmake/**'
- 'sherpa-onnx/csrc/*'
@@ -70,6 +72,14 @@ jobs:
ls -lh ./bin/Release/sherpa-onnx.exe
- name: Test Audio tagging
shell: bash
run: |
export PATH=$PWD/build/bin/Release:$PATH
export EXE=sherpa-onnx-offline-audio-tagging.exe
.github/scripts/test-audio-tagging.sh
- name: Test C API
shell: bash
run: |

View File

@@ -14,6 +14,7 @@ on:
- '.github/scripts/test-offline-ctc.sh'
- '.github/scripts/test-offline-tts.sh'
- '.github/scripts/test-online-ctc.sh'
- '.github/scripts/test-audio-tagging.sh'
- 'CMakeLists.txt'
- 'cmake/**'
- 'sherpa-onnx/csrc/*'
@@ -28,6 +29,7 @@ on:
- '.github/scripts/test-offline-ctc.sh'
- '.github/scripts/test-offline-tts.sh'
- '.github/scripts/test-online-ctc.sh'
- '.github/scripts/test-audio-tagging.sh'
- 'CMakeLists.txt'
- 'cmake/**'
- 'sherpa-onnx/csrc/*'
@@ -85,6 +87,13 @@ jobs:
# export EXE=sherpa-onnx-offline-language-identification.exe
#
# .github/scripts/test-spoken-language-identification.sh
- name: Test Audio tagging
shell: bash
run: |
export PATH=$PWD/build/bin/Release:$PATH
export EXE=sherpa-onnx-offline-audio-tagging.exe
.github/scripts/test-audio-tagging.sh
- name: Test online CTC
shell: bash