Add Lazarus example for generating subtitles using Silero VAD with non-streaming ASR (#1251)
This commit is contained in:
Fangjun Kuang
81
scripts/lazarus/build-generate-subtitles.sh.in
Normal file
81
scripts/lazarus/build-generate-subtitles.sh.in
Normal file
@@ -0,0 +1,81 @@
|
||||
#!/usr/bin/env bash
|
||||
#
|
||||
# It expects that there are 4 directories inside /tmp
|
||||
#
|
||||
# macos-x64
|
||||
# macos-arm64
|
||||
# linux-x64
|
||||
# windows-x64
|
||||
#
|
||||
# Generated files are saved in /tmp/out/*.tar.bz2
|
||||
|
||||
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]}) $*"
|
||||
}
|
||||
|
||||
SHERPA_ONNX_VERSION=$(grep "SHERPA_ONNX_VERSION" ./CMakeLists.txt | cut -d " " -f 2 | cut -d '"' -f 2)
|
||||
|
||||
curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/silero_vad.onnx
|
||||
|
||||
os_array=(
|
||||
linux-x64
|
||||
macos-x64
|
||||
macos-arm64
|
||||
windows-x64
|
||||
)
|
||||
|
||||
for os in ${os_array[@]}; do
|
||||
if [[ $os == macos-x64 || $os == macos-arm64 ]]; then
|
||||
cp -v silero_vad.onnx /tmp/$os/Contents/Resources/
|
||||
else
|
||||
cp -v silero_vad.onnx /tmp/$os/
|
||||
fi
|
||||
done
|
||||
|
||||
{% for model in model_list %}
|
||||
model_name={{ model.model_name }}
|
||||
lang={{ model.lang }}
|
||||
short_name={{ model.short_name }}
|
||||
|
||||
curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/${model_name}.tar.bz2
|
||||
tar xvf ${model_name}.tar.bz2
|
||||
rm ${model_name}.tar.bz2
|
||||
|
||||
{{ model.cmd }}
|
||||
|
||||
ls -lh $model_name
|
||||
|
||||
for os in ${os_array[@]}; do
|
||||
dst=sherpa-onnx-$SHERPA_ONNX_VERSION-generate-subtitles-$os-$short_name-$lang
|
||||
src=/tmp/$os
|
||||
|
||||
cp -a $src $dst
|
||||
d=$dst
|
||||
|
||||
if [[ $os == macos-x64 || $os == macos-arm64 ]]; then
|
||||
mv $dst $dst.app
|
||||
|
||||
dst=$dst.app
|
||||
d=$dst/Contents/Resources
|
||||
fi
|
||||
|
||||
cp -v $model_name/*.onnx $d
|
||||
cp -v $model_name/tokens.txt $d
|
||||
|
||||
d=$dst
|
||||
|
||||
tar cjvf $d.tar.bz2 $d
|
||||
|
||||
ls -lh
|
||||
|
||||
mkdir -p /tmp/out
|
||||
mv $d.tar.bz2 /tmp/out
|
||||
ls -lh /tmp/out
|
||||
done
|
||||
|
||||
rm -rf $model_name
|
||||
{% endfor %}
|
||||
176
scripts/lazarus/generate-subtitles.py
Executable file
176
scripts/lazarus/generate-subtitles.py
Executable file
@@ -0,0 +1,176 @@
|
||||
#!/usr/bin/env python3
|
||||
|
||||
import argparse
|
||||
from dataclasses import dataclass
|
||||
from typing import List, Optional
|
||||
|
||||
import jinja2
|
||||
|
||||
|
||||
def get_args():
|
||||
parser = argparse.ArgumentParser()
|
||||
parser.add_argument(
|
||||
"--total",
|
||||
type=int,
|
||||
default=1,
|
||||
help="Number of runners",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--index",
|
||||
type=int,
|
||||
default=0,
|
||||
help="Index of the current runner",
|
||||
)
|
||||
return parser.parse_args()
|
||||
|
||||
|
||||
@dataclass
|
||||
class Model:
|
||||
model_name: str
|
||||
lang: str
|
||||
short_name: str = ""
|
||||
cmd: str = ""
|
||||
|
||||
|
||||
def get_models():
|
||||
models = [
|
||||
Model(
|
||||
model_name="sherpa-onnx-whisper-tiny.en",
|
||||
lang="en",
|
||||
short_name="whisper_tiny.en",
|
||||
cmd="""
|
||||
pushd $model_name
|
||||
rm -fv tiny.en-encoder.onnx
|
||||
rm -fv tiny.en-decoder.onnx
|
||||
|
||||
mv -v tiny.en-encoder.int8.onnx whisper-encoder.onnx
|
||||
mv -v tiny.en-decoder.int8.onnx whisper-decoder.onnx
|
||||
mv -v tiny.en-tokens.txt tokens.txt
|
||||
|
||||
popd
|
||||
""",
|
||||
),
|
||||
Model(
|
||||
model_name="sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17",
|
||||
lang="zh_en_ko_ja_yue",
|
||||
short_name="sense_voice",
|
||||
cmd="""
|
||||
pushd $model_name
|
||||
rm -fv model.onnx
|
||||
mv -v model.int8.onnx sense-voice.onnx
|
||||
popd
|
||||
""",
|
||||
),
|
||||
Model(
|
||||
model_name="sherpa-onnx-paraformer-zh-2023-09-14",
|
||||
lang="zh_en",
|
||||
short_name="paraformer_2023_09_14",
|
||||
cmd="""
|
||||
pushd $model_name
|
||||
rm -fv model.onnx
|
||||
mv -v model.int8.onnx paraformer.onnx
|
||||
popd
|
||||
""",
|
||||
),
|
||||
Model(
|
||||
model_name="sherpa-onnx-paraformer-zh-small-2024-03-09",
|
||||
lang="zh_en",
|
||||
short_name="paraformer_small_2024_03_09",
|
||||
cmd="""
|
||||
pushd $model_name
|
||||
rm -fv model.onnx
|
||||
mv -v model.int8.onnx paraformer.onnx
|
||||
popd
|
||||
""",
|
||||
),
|
||||
Model(
|
||||
model_name="sherpa-onnx-zipformer-gigaspeech-2023-12-12",
|
||||
lang="en",
|
||||
short_name="zipformer_gigaspeech_2023_12_12",
|
||||
cmd="""
|
||||
pushd $model_name
|
||||
mv encoder-epoch-30-avg-1.int8.onnx transducer-encoder.onnx
|
||||
mv decoder-epoch-30-avg-1.onnx transducer-decoder.onnx
|
||||
mv joiner-epoch-30-avg-1.int8.onnx transducer-joiner.onnx
|
||||
|
||||
rm -fv encoder-epoch-30-avg-1.onnx
|
||||
rm -fv decoder-epoch-30-avg-1.int8.onnx
|
||||
rm -fv joiner-epoch-30-avg-1.onnx
|
||||
|
||||
popd
|
||||
""",
|
||||
),
|
||||
Model(
|
||||
model_name="icefall-asr-zipformer-wenetspeech-20230615",
|
||||
lang="zh",
|
||||
short_name="zipformer_wenetspeech",
|
||||
cmd="""
|
||||
pushd $model_name
|
||||
|
||||
rm -rfv test_wavs
|
||||
rm -fv README.md
|
||||
mv -v data/lang_char/tokens.txt ./
|
||||
rm -rfv data/lang_char
|
||||
|
||||
mv -v exp/encoder-epoch-12-avg-4.int8.onnx ./
|
||||
mv -v exp/decoder-epoch-12-avg-4.onnx ./
|
||||
mv -v exp/joiner-epoch-12-avg-4.int8.onnx ./
|
||||
rm -rfv exp
|
||||
|
||||
mv -v encoder-epoch-12-avg-4.int8.onnx transducer-encoder.onnx
|
||||
mv -v decoder-epoch-12-avg-4.onnx transducer-decoder.onnx
|
||||
mv -v joiner-epoch-12-avg-4.int8.onnx transducer-joiner.onnx
|
||||
|
||||
ls -lh
|
||||
|
||||
popd
|
||||
""",
|
||||
),
|
||||
]
|
||||
return models
|
||||
|
||||
|
||||
def main():
|
||||
args = get_args()
|
||||
index = args.index
|
||||
total = args.total
|
||||
assert 0 <= index < total, (index, total)
|
||||
|
||||
all_model_list = get_models()
|
||||
|
||||
num_models = len(all_model_list)
|
||||
|
||||
num_per_runner = num_models // total
|
||||
if num_per_runner <= 0:
|
||||
raise ValueError(f"num_models: {num_models}, num_runners: {total}")
|
||||
|
||||
start = index * num_per_runner
|
||||
end = start + num_per_runner
|
||||
|
||||
remaining = num_models - args.total * num_per_runner
|
||||
|
||||
print(f"{index}/{total}: {start}-{end}/{num_models}")
|
||||
|
||||
d = dict()
|
||||
d["model_list"] = all_model_list[start:end]
|
||||
if index < remaining:
|
||||
s = args.total * num_per_runner + index
|
||||
d["model_list"].append(all_model_list[s])
|
||||
print(f"{s}/{num_models}")
|
||||
|
||||
filename_list = [
|
||||
"./build-generate-subtitles.sh",
|
||||
]
|
||||
for filename in filename_list:
|
||||
environment = jinja2.Environment()
|
||||
with open(f"{filename}.in") as f:
|
||||
s = f.read()
|
||||
template = environment.from_string(s)
|
||||
|
||||
s = template.render(**d)
|
||||
with open(filename, "w") as f:
|
||||
print(s, file=f)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
Reference in New Issue
Block a user