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
Dev gzf new (#1551)
zhifu gao16c70d831daa
diff --git a/funasr/train_utils/trainer.py b/funasr/train_utils/trainer.pyindex 8fd97bff7..55dce9996 100644--- a/funasr/train_utils/trainer.py+++ b/funasr/train_utils/trainer.py@@ -524,7 +524,7 @@ class Trainer:f"step: {batch_idx + 1}/{batch_num_epoch}, total step: {self.batch_total}, "f"(loss_avg_rank: {loss:.3f}), "f"(loss_avg_epoch: {loss_avg_epoch:.3f}), "- f"(ppl_avg_epoch: {math.exp(loss_avg_epoch):.3f}), "+ f"(ppl_avg_epoch: {math.exp(loss_avg_epoch):.3e}), "f"(acc_avg_epoch: {acc_avg_epoch:.3f}), "f"(lr: {lr:.3e}), "f"{[(k, round(v.detach().cpu().item(), 3)) for k, v in stats.items()]}, "