EngineX-Iluvatar/enginex-mr_series-sherpa-onnx
Archived
16
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Add HLG decoding for streaming CTC models (#731)

Browse Source
This commit is contained in:
Fangjun Kuang
2024-04-03 21:31:42 +08:00
committed by GitHub
parent f8832cb5f2
commit db67e00c77
28 changed files with 668 additions and 82 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
.github/scripts/test-online-ctc.sh vendored
View File

@@ -1,6 +1,6 @@
#!/usr/bin/env bash
set -e
set -ex
log() {
# This function is from espnet
@@ -13,6 +13,26 @@ echo "PATH: $PATH"
which $EXE
log "------------------------------------------------------------"
log "Run streaming Zipformer2 CTC HLG decoding "
log "------------------------------------------------------------"
curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18.tar.bz2
tar xvf sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18.tar.bz2
rm sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18.tar.bz2
repo=$PWD/sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18
ls -lh $repo
echo "pwd: $PWD"
$EXE \
--zipformer2-ctc-model=$repo/ctc-epoch-30-avg-3-chunk-16-left-128.int8.onnx \
--ctc-graph=$repo/HLG.fst \
--tokens=$repo/tokens.txt \
$repo/test_wavs/0.wav \
$repo/test_wavs/1.wav \
$repo/test_wavs/8k.wav
rm -rf sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18
log "------------------------------------------------------------"
log "Run streaming Zipformer2 CTC "
log "------------------------------------------------------------"

19
.github/scripts/test-python.sh vendored
View File

@@ -1,6 +1,6 @@
#!/usr/bin/env bash
set -e
set -ex
log() {
# This function is from espnet
@@ -8,6 +8,23 @@ log() {
echo -e "$(date '+%Y-%m-%d %H:%M:%S') (${fname}:${BASH_LINENO[0]}:${FUNCNAME[1]}) $*"
}
log "test streaming zipformer2 ctc HLG decoding"
curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18.tar.bz2
tar xvf sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18.tar.bz2
rm sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18.tar.bz2
repo=sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18
python3 ./python-api-examples/online-zipformer-ctc-hlg-decode-file.py \
--debug 1 \
--tokens ./sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18/tokens.txt \
--graph ./sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18/HLG.fst \
--model ./sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18/ctc-epoch-30-avg-3-chunk-16-left-128.int8.onnx \
./sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18/test_wavs/0.wav
rm -rf sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18
mkdir -p /tmp/icefall-models
dir=/tmp/icefall-models

15
.github/workflows/linux.yaml vendored
View File

@@ -124,6 +124,14 @@ jobs:
name: release-${{ matrix.build_type }}-with-shared-lib-${{ matrix.shared_lib }}-with-tts-${{ matrix.with_tts }}
path: build/bin/*
- name: Test online CTC
shell: bash
run: |
export PATH=$PWD/build/bin:$PATH
export EXE=sherpa-onnx
.github/scripts/test-online-ctc.sh
- name: Test C API
shell: bash
run: |
@@ -149,13 +157,6 @@ jobs:
.github/scripts/test-kws.sh
- name: Test online CTC
shell: bash
run: |
export PATH=$PWD/build/bin:$PATH
export EXE=sherpa-onnx
.github/scripts/test-online-ctc.sh
- name: Test offline Whisper
if: matrix.build_type != 'Debug'