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
feat: speed up fbank's lfr (#2246)
Tang Linjie8b1be8c3cba8
diff --git a/funasr/frontends/wav_frontend.py b/funasr/frontends/wav_frontend.pyindex 332420898..da23f9c7c 100644--- a/funasr/frontends/wav_frontend.py+++ b/funasr/frontends/wav_frontend.py@@ -62,17 +62,16 @@ def apply_lfr(inputs, lfr_m, lfr_n):left_padding = inputs[0].repeat((lfr_m - 1) // 2, 1)inputs = torch.vstack((left_padding, inputs))T = T + (lfr_m - 1) // 2- for i in range(T_lfr):- if lfr_m <= T - i * lfr_n:- LFR_inputs.append((inputs[i * lfr_n : i * lfr_n + lfr_m]).view(1, -1))- else: # process last LFR frame- num_padding = lfr_m - (T - i * lfr_n)- frame = (inputs[i * lfr_n :]).view(-1)- for _ in range(num_padding):- frame = torch.hstack((frame, inputs[-1]))- LFR_inputs.append(frame)- LFR_outputs = torch.vstack(LFR_inputs)- return LFR_outputs.type(torch.float32)+ feat_dim = inputs.shape[-1]+ strides = (lfr_n * feat_dim, 1)+ sizes = (T_lfr, lfr_m * feat_dim)+ last_idx = (T - lfr_m) // lfr_n + 1+ num_padding = lfr_m - (T - last_idx * lfr_n)+ if num_padding > 0:+ num_padding = (2 * lfr_m - 2 * T + (T_lfr - 1 + last_idx) * lfr_n) / 2 * (T_lfr - last_idx)+ inputs = torch.vstack([inputs] + [inputs[-1:]] * int(num_padding))+ LFR_outputs = inputs.as_strided(sizes, strides)+ return LFR_outputs.clone().type(torch.float32)@tables.register("frontend_classes", "wav_frontend")@@ -289,24 +288,24 @@ class WavFrontendOnline(nn.Module):np.ceil((T - (lfr_m - 1) // 2) / lfr_n)) # minus the right context: (lfr_m - 1) // 2splice_idx = T_lfr- for i in range(T_lfr):- if lfr_m <= T - i * lfr_n:- LFR_inputs.append((inputs[i * lfr_n : i * lfr_n + lfr_m]).view(1, -1))- else: # process last LFR frame- if is_final:- num_padding = lfr_m - (T - i * lfr_n)- frame = (inputs[i * lfr_n :]).view(-1)- for _ in range(num_padding):- frame = torch.hstack((frame, inputs[-1]))- LFR_inputs.append(frame)- else:- # update splice_idx and break the circle- splice_idx = i- break+ feat_dim = inputs.shape[-1]+ ori_inputs = inputs+ strides = (lfr_n * feat_dim, 1)+ sizes = (T_lfr, lfr_m * feat_dim)+ last_idx = (T - lfr_m) // lfr_n + 1+ num_padding = lfr_m - (T - last_idx * lfr_n)+ if is_final:+ if num_padding > 0:+ num_padding = (2 * lfr_m - 2 * T + (T_lfr - 1 + last_idx) * lfr_n) / 2 * (T_lfr - last_idx)+ inputs = torch.vstack([inputs] + [inputs[-1:]] * int(num_padding))+ else:+ if num_padding > 0:+ sizes = (last_idx, lfr_m * feat_dim)+ splice_idx = last_idxsplice_idx = min(T - 1, splice_idx * lfr_n)- lfr_splice_cache = inputs[splice_idx:, :]- LFR_outputs = torch.vstack(LFR_inputs)- return LFR_outputs.type(torch.float32), lfr_splice_cache, splice_idx+ LFR_outputs = inputs[:splice_idx].as_strided(sizes, strides)+ lfr_splice_cache = ori_inputs[splice_idx:, :]+ return LFR_outputs.clone().type(torch.float32), lfr_splice_cache, splice_idx@staticmethoddef compute_frame_num(