Add Python API examples for speaker recognition with VAD and ASR.

[csukuangfj]

Example 1: Speaker recognition + VAD

python3 ./python-api-examples/speaker-identification-with-vad.py \
  --silero-vad-model ./silero_vad.onnx \
  --threshold 0.5 \
  --model ./3dspeaker_speech_eres2net_large_sv_zh-cn_3dspeaker_16k.onnx \
  --speaker-file ./speaker.txt

Example 2: Speaker recognition + VAD + non-streaming ASR

python3 ./python-api-examples/speaker-identification-with-vad-non-streaming-asr.py \
  --paraformer ./sherpa-onnx-paraformer-zh-2023-03-28/model.int8.onnx \
  --tokens ./sherpa-onnx-paraformer-zh-2023-03-28/tokens.txt \
  --silero-vad-model ./silero_vad.onnx \
  --threshold 0.5 \
  --model ./3dspeaker_speech_eres2net_large_sv_zh-cn_3dspeaker_16k.onnx \
  --speaker-file ./speaker.txt

where speaker.txt contains:

雷军 ./sr-data/enroll/leijun-sr-1.wav
雷军 ./sr-data/enroll/leijun-sr-2.wav
fangjun ./sr-data/enroll/fangjun-sr-1.wav
fangjun ./sr-data/enroll/fangjun-sr-2.wav
fangjun ./sr-data/enroll/fangjun-sr-3.wav
刘德华 ./sr-data/enroll/liudehua-sr-1.wav
刘德华 ./sr-data/enroll/liudehua-sr-2.wav

You can find the test data sr-data at
https://github.com/csukuangfj/sr-data