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
auto frontend
游雁a0becf3cd81f
diff --git a/funasr/datasets/openai_datasets/datasets.py b/funasr/datasets/openai_datasets/datasets.pyindex 73e9ec0ac..f6127b65a 100644--- a/funasr/datasets/openai_datasets/datasets.py+++ b/funasr/datasets/openai_datasets/datasets.py@@ -215,7 +215,7 @@ class OpenAIDataset(torch.utils.data.Dataset):if b < 2:beg = 0logging.info(- f"Warning, b * t: {b * t} > {self.batch_size}, drop half data {i}th, beg:{beg}"+ f"Warning, b * t: {b * t} > {self.batch_size}, b: {b}, t: {t}, drop half data {i}th, beg:{beg}")for key, data_list in outputs.items():outputs[key] = outputs[key][beg : beg + b : 2]