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
update train recipe
游雁7357bb6c45eb
diff --git a/funasr/train_utils/trainer.py b/funasr/train_utils/trainer.pyindex f4ef77379..d24023d21 100644--- a/funasr/train_utils/trainer.py+++ b/funasr/train_utils/trainer.py@@ -157,7 +157,7 @@ class Trainer:self._resume_checkpoint(self.output_dir)for epoch in range(self.start_epoch, self.max_epoch + 1):-+ time1 = time.perf_counter()self._train_epoch(epoch)@@ -179,6 +179,9 @@ class Trainer:self.scheduler.step()+ time2 = time.perf_counter()+ time_escaped = (time2 - time1)/3600.0+ print(f"time_escaped_epoch: {time_escaped:.3f} hours, estimated to finish: {(self.max_epoch-epoch)*time_escaped:.3f}")if self.rank == 0:average_checkpoints(self.output_dir, self.avg_nbest_model)@@ -285,7 +288,7 @@ class Trainer:)lr = self.scheduler.get_last_lr()[0]time_now = datetime.now()- time_now = now.strftime("%Y-%m-%d %H:%M:%S")+ time_now = time_now.strftime("%Y-%m-%d %H:%M:%S")description = (f"{time_now}, "f"rank: {self.local_rank}, "