Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
"""Measure batched GPU runtime RTF without silently benchmarking CPU fallback."""
def main(argv=None):
if __package__:
from .runtime_eval import benchmark, parse_args
else:
from runtime_eval import benchmark, parse_args
benchmark(parse_args(argv, gpu=True), batched=True)
if __name__ == "__main__":
main()