Merge branch 'master-test0109'

# Conflicts: # ios-swift/.gitignore
XiaYucca · Jan 9, 2025 · f5dbf84 · f5dbf84
2 parents 748892f + 47c19da
commit f5dbf84
Show file tree

Hide file tree

Showing 436 changed files with 14,307 additions and 1,186 deletions.
diff --git a/.github/scripts/test-cxx-api.sh b/.github/scripts/test-cxx-api.sh
@@ -0,0 +1,42 @@
+#!/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 "CXX_STREAMING_ZIPFORMER_EXE is $CXX_STREAMING_ZIPFORMER_EXE"
+echo "CXX_WHISPER_EXE is $CXX_WHISPER_EXE"
+echo "CXX_SENSE_VOICE_EXE is $CXX_SENSE_VOICE_EXE"
+echo "PATH: $PATH"
+
+log "------------------------------------------------------------"
+log "Test streaming zipformer CXX API"
+log "------------------------------------------------------------"
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20.tar.bz2
+tar xvf sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20.tar.bz2
+rm sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20.tar.bz2
+$CXX_STREAMING_ZIPFORMER_EXE
+rm -rf sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20
+
+log "------------------------------------------------------------"
+log "Test Whisper CXX API"
+log "------------------------------------------------------------"
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-whisper-tiny.en.tar.bz2
+tar xvf sherpa-onnx-whisper-tiny.en.tar.bz2
+rm sherpa-onnx-whisper-tiny.en.tar.bz2
+$CXX_WHISPER_EXE
+rm -rf sherpa-onnx-whisper-tiny.en
+
+log "------------------------------------------------------------"
+log "Test SenseVoice CXX API"
+log "------------------------------------------------------------"
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17.tar.bz2
+tar xvf sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17.tar.bz2
+rm sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17.tar.bz2
+
+$CXX_SENSE_VOICE_EXE
+rm -rf sherpa-onnx-sense-voice-*
diff --git a/.github/scripts/test-dart.sh b/.github/scripts/test-dart.sh
@@ -36,6 +36,10 @@ echo "----zipformer transducer----"
 ./run-zipformer-transducer.sh
 rm -rf sherpa-onnx-*
 
+echo "----moonshine----"
+./run-moonshine.sh
+rm -rf sherpa-onnx-*
+
 echo "----whisper----"
 ./run-whisper.sh
 rm -rf sherpa-onnx-*
@@ -77,6 +81,10 @@ echo '----------TeleSpeech CTC----------'
 ./run-telespeech-ctc.sh
 rm -rf sherpa-onnx-*
 
+echo '----------moonshine----------'
+./run-moonshine.sh
+rm -rf sherpa-onnx-*
+
 echo '----------whisper----------'
 ./run-whisper.sh
 rm -rf sherpa-onnx-*

diff --git a/.github/scripts/test-dot-net.sh b/.github/scripts/test-dot-net.sh
@@ -9,6 +9,9 @@ rm -fv *.wav
 rm -rfv sherpa-onnx-pyannote-*
 
 cd ../offline-decode-files
+./run-moonshine.sh
+rm -rf sherpa-onnx-*
+
 ./run-sense-voice-ctc.sh
 rm -rf sherpa-onnx-*
 

diff --git a/.github/scripts/test-nodejs-addon-npm.sh b/.github/scripts/test-nodejs-addon-npm.sh
@@ -10,6 +10,19 @@ arch=$(node -p "require('os').arch()")
 platform=$(node -p "require('os').platform()")
 node_version=$(node -p "process.versions.node.split('.')[0]")
 
+echo "----------non-streaming asr moonshine + vad----------"
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+tar xvf sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+rm sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/Obama.wav
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/silero_vad.onnx
+
+node ./test_vad_with_non_streaming_asr_moonshine.js
+rm -rf sherpa-onnx-*
+rm *.wav
+rm *.onnx
+
 echo "----------non-streaming speaker diarization----------"
 
 curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/speaker-segmentation-models/sherpa-onnx-pyannote-segmentation-3-0.tar.bz2
@@ -24,7 +37,7 @@ node ./test_offline_speaker_diarization.js
 
 rm -rfv *.onnx *.wav sherpa-onnx-pyannote-*
 
-echo "----------non-streaming asr + vad----------"
+echo "----------non-streaming asr whisper + vad----------"
 curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-whisper-tiny.en.tar.bz2
 tar xvf sherpa-onnx-whisper-tiny.en.tar.bz2
 rm sherpa-onnx-whisper-tiny.en.tar.bz2
@@ -218,6 +231,11 @@ rm sherpa-onnx-whisper-tiny.en.tar.bz2
 node ./test_asr_non_streaming_whisper.js
 rm -rf sherpa-onnx-whisper-tiny.en
 
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+tar xvf sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+rm sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
 
+node ./test_asr_non_streaming_moonshine.js
+rm -rf sherpa-onnx-*
 
 ls -lh
diff --git a/.github/scripts/test-nodejs-npm.sh b/.github/scripts/test-nodejs-npm.sh
@@ -21,6 +21,23 @@ curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/speaker-segm
 node ./test-offline-speaker-diarization.js
 rm -rfv *.wav *.onnx sherpa-onnx-pyannote-*
 
+echo '-----vad+moonshine----------'
+
+curl -LS -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-whisper-tiny.en.tar.bz2
+tar xvf sherpa-onnx-whisper-tiny.en.tar.bz2
+rm sherpa-onnx-whisper-tiny.en.tar.bz2
+
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+tar xvf sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+rm sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/Obama.wav
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/silero_vad.onnx
+node ./test-vad-with-non-streaming-asr-whisper.js
+rm Obama.wav
+rm silero_vad.onnx
+rm -rf sherpa-onnx-moonshine-*
+
 echo '-----vad+whisper----------'
 
 curl -LS -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-whisper-tiny.en.tar.bz2
@@ -90,6 +107,13 @@ rm sherpa-onnx-whisper-tiny.en.tar.bz2
 node ./test-offline-whisper.js
 rm -rf sherpa-onnx-whisper-tiny.en
 
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+tar xvf sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+rm sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+
+node ./test-offline-moonshine.js
+rm -rf sherpa-onnx-moonshine-*
+
 # online asr
 curl -LS -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-paraformer-bilingual-zh-en.tar.bz2
 tar xvf sherpa-onnx-streaming-paraformer-bilingual-zh-en.tar.bz2

diff --git a/.github/scripts/test-offline-ctc.sh b/.github/scripts/test-offline-ctc.sh
@@ -15,6 +15,21 @@ echo "PATH: $PATH"
 
 which $EXE
 
+log "------------------------------------------------------------"
+log "Run NeMo GigaAM Russian models"
+log "------------------------------------------------------------"
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-nemo-ctc-giga-am-russian-2024-10-24.tar.bz2
+tar xvf sherpa-onnx-nemo-ctc-giga-am-russian-2024-10-24.tar.bz2
+rm sherpa-onnx-nemo-ctc-giga-am-russian-2024-10-24.tar.bz2
+
+$EXE \
+  --nemo-ctc-model=./sherpa-onnx-nemo-ctc-giga-am-russian-2024-10-24/model.int8.onnx \
+  --tokens=./sherpa-onnx-nemo-ctc-giga-am-russian-2024-10-24/tokens.txt \
+  --debug=1 \
+  ./sherpa-onnx-nemo-ctc-giga-am-russian-2024-10-24/test_wavs/example.wav
+
+rm -rf sherpa-onnx-nemo-ctc-giga-am-russian-2024-10-24
+
 log "------------------------------------------------------------"
 log "Run SenseVoice models"
 log "------------------------------------------------------------"

diff --git a/.github/scripts/test-offline-moonshine.sh b/.github/scripts/test-offline-moonshine.sh
@@ -0,0 +1,50 @@
+#!/usr/bin/env bash
+
+set -e
+
+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]}) $*"
+}
+
+export GIT_CLONE_PROTECTION_ACTIVE=false
+
+echo "EXE is $EXE"
+echo "PATH: $PATH"
+
+which $EXE
+
+names=(
+tiny
+base
+)
+
+for name in ${names[@]}; do
+  log "------------------------------------------------------------"
+  log "Run $name"
+  log "------------------------------------------------------------"
+
+  repo_url=https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-whisper-$name.tar.bz2
+  repo_url=https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-moonshine-$name-en-int8.tar.bz2
+  curl -SL -O $repo_url
+  tar xvf sherpa-onnx-moonshine-$name-en-int8.tar.bz2
+  rm sherpa-onnx-moonshine-$name-en-int8.tar.bz2
+  repo=sherpa-onnx-moonshine-$name-en-int8
+  log "Start testing ${repo_url}"
+
+  log "test int8 onnx"
+
+  time $EXE \
+    --moonshine-preprocessor=$repo/preprocess.onnx \
+    --moonshine-encoder=$repo/encode.int8.onnx \
+    --moonshine-uncached-decoder=$repo/uncached_decode.int8.onnx \
+    --moonshine-cached-decoder=$repo/cached_decode.int8.onnx \
+    --tokens=$repo/tokens.txt \
+    --num-threads=2 \
+    $repo/test_wavs/0.wav \
+    $repo/test_wavs/1.wav \
+    $repo/test_wavs/8k.wav
+
+  rm -rf $repo
+done
diff --git a/.github/scripts/test-python.sh b/.github/scripts/test-python.sh
@@ -8,6 +8,16 @@ log() {
   echo -e "$(date '+%Y-%m-%d %H:%M:%S') (${fname}:${BASH_LINENO[0]}:${FUNCNAME[1]}) $*"
 }
 
+log "test offline Moonshine"
+
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+tar xvf sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+rm sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
+
+python3 ./python-api-examples/offline-moonshine-decode-files.py
+
+rm -rf sherpa-onnx-moonshine-tiny-en-int8
+
 log "test offline speaker diarization"
 
 curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/speaker-segmentation-models/sherpa-onnx-pyannote-segmentation-3-0.tar.bz2