Add fast beam search decoding

egs/librispeech/ASR/pruned_transducer_stateless/beam_search.py

@@ -17,9 +17,86 @@
 from dataclasses import dataclass
 from typing import Dict, List, Optional
 
+import k2
 import torch
 from model import Transducer
 
+from icefall.decode import one_best_decoding
+from icefall.utils import get_texts
+
+
+def fast_beam_search(
+    decoding_graph: k2.Fsa,
+    model: Transducer,
+    encoder_out: torch.Tensor,
+    encoder_out_lens: torch.Tensor,
+    beam: float,
+    max_states: int,
+    max_contexts: int,
+) -> List[int]:
+    """It limits the maximum number of symbols per frame to 1.
+
+    Args:
+      model:
+        An instance of `Transducer`.
+      encoder_out:
+        A tensor of shape (N, T, C) from the encoder. Support only N==1 for now.
+      decoding_graph:
+        Decoding graph used for decoding, may be a TrivialGraph or a HLG.
+      beam:
+        Beam value, similar to the beam used in Kaldi..
+      max_states:
+        Max states per stream per frame.
+      max_contexts:
+        Max contexts pre stream per frame.
+    Returns:
+      Return the decoded result.
+    """
+    assert encoder_out.ndim == 3
+
+    context_size = model.decoder.context_size
+    vocab_size = model.decoder.vocab_size
+
+    B, T, C = encoder_out.shape
+
+    config = k2.RnntDecodingConfig(
+        vocab_size=vocab_size,
+        decoder_history_len=context_size,
+        beam=beam,
+        max_contexts=max_contexts,
+        max_states=max_states,
+    )
+    indivisual_streams = []
+    for i in range(B):
+        indivisual_streams.append(k2.RnntDecodingStream(decoding_graph))
+    decoding_streams = k2.RnntDecodingStreams(indivisual_streams, config)
+
+    for t in range(T):
+        # shape is a RaggedShape of shape (B, context)
+        # contexts is a Tensor of shape (shape.NumElements(), context_size)
+        shape, contexts = decoding_streams.get_contexts()
+        # decoder_out is of shape (shape.NumElements(), 1, decoder_out_dim)
+        decoder_out = model.decoder(contexts, need_pad=False)
+        # current_encoder_out is of shape
+        # (shape.NumElements(), 1, encoder_out_dim)
+        # fmt: off
+        current_encoder_out = torch.index_select(
+            encoder_out[:, t:t + 1, :], 0, shape.row_ids(1)
+        )
+        # fmt: on
+        logits = model.joiner(
+            current_encoder_out.unsqueeze(2), decoder_out.unsqueeze(1)
+        )
+        logits = logits.squeeze(1).squeeze(1)
+        log_probs = logits.log_softmax(dim=-1)
+        decoding_streams.advance(log_probs)
+    decoding_streams.detach()
+    lattice = decoding_streams.format_output(encoder_out_lens.tolist())
+
+    best_path = one_best_decoding(lattice)
+    hyps = get_texts(best_path)
+    return hyps
+
 
 def greedy_search(
     model: Transducer, encoder_out: torch.Tensor, max_sym_per_frame: int

egs/librispeech/ASR/pruned_transducer_stateless/decode.py

@@ -42,6 +42,17 @@ Usage:
         --max-duration 100 \
         --decoding-method modified_beam_search \
         --beam-size 4
+
+(4) fast beam search
+./pruned_transducer_stateless/decode.py \
+        --epoch 28 \
+        --avg 15 \
+        --exp-dir ./pruned_transducer_stateless/exp \
+        --max-duration 1500 \
+        --decoding-method fast_beam_search \
+        --beam 8 \
+        --max-contexts 10 \
+        --max-states 20
 """
 
 
@@ -51,11 +62,17 @@ from collections import defaultdict
 from pathlib import Path
 from typing import Dict, List, Tuple
 
+import k2
 import sentencepiece as spm
 import torch
 import torch.nn as nn
 from asr_datamodule import LibriSpeechAsrDataModule
-from beam_search import beam_search, greedy_search, modified_beam_search
+from beam_search import (
+    beam_search,
+    fast_beam_search,
+    greedy_search,
+    modified_beam_search,
+)
 from train import get_params, get_transducer_model
 
 from icefall.checkpoint import average_checkpoints, load_checkpoint
@@ -110,6 +127,7 @@ def get_parser():
           - greedy_search
           - beam_search
           - modified_beam_search
+          - fast_beam_search
         """,
     )
 
@@ -121,6 +139,30 @@ def get_parser():
         beam_search or modified_beam_search""",
     )
 
+    parser.add_argument(
+        "--beam",
+        type=float,
+        default=8,
+        help="""Used only when --decoding-method is
+        fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--max-contexts",
+        type=int,
+        default=5,
+        help="""Used only when --decoding-method is
+        fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--max-states",
+        type=int,
+        default=10,
+        help="""Used only when --decoding-method is
+        fast_beam_search""",
+    )
+
     parser.add_argument(
         "--context-size",
         type=int,
@@ -143,6 +185,7 @@ def decode_one_batch(
     params: AttributeDict,
     model: nn.Module,
     sp: spm.SentencePieceProcessor,
+    decoding_graph: k2.Fsa,
     batch: dict,
 ) -> Dict[str, List[List[str]]]:
     """Decode one batch and return the result in a dict. The dict has the
@@ -162,6 +205,9 @@ def decode_one_batch(
         The neural model.
       sp:
         The BPE model.
+      decoding_graph:
+        The decoding graph. Can be either a `k2.TrivialGraph` or HLG, Used
+        only when --decoding_method is fast_beam_search.
       batch:
         It is the return value from iterating
         `lhotse.dataset.K2SpeechRecognitionDataset`. See its documentation
@@ -184,36 +230,62 @@ def decode_one_batch(
         x=feature, x_lens=feature_lens
     )
     hyps = []
-    batch_size = encoder_out.size(0)
 
-    for i in range(batch_size):
+    if params.decoding_method == "fast_beam_search":
-        # fmt: off
+        hyp_tokens = fast_beam_search(
-        encoder_out_i = encoder_out[i:i+1, :encoder_out_lens[i]]
+            model=model,
-        # fmt: on
+            decoding_graph=decoding_graph,
-        if params.decoding_method == "greedy_search":
+            encoder_out=encoder_out,
-            hyp = greedy_search(
+            encoder_out_lens=encoder_out_lens,
-                model=model,
+            beam=params.beam,
-                encoder_out=encoder_out_i,
+            max_contexts=params.max_contexts,
-                max_sym_per_frame=params.max_sym_per_frame,
+            max_states=params.max_states,
-            )
+        )
-        elif params.decoding_method == "beam_search":
+        for hyp in sp.decode(hyp_tokens):
-            hyp = beam_search(
+            hyps.append(hyp.split())
-                model=model, encoder_out=encoder_out_i, beam=params.beam_size
+    else:
-            )
+        batch_size = encoder_out.size(0)
-        elif params.decoding_method == "modified_beam_search":
+
-            hyp = modified_beam_search(
+        for i in range(batch_size):
-                model=model, encoder_out=encoder_out_i, beam=params.beam_size
+            # fmt: off
-            )
+            encoder_out_i = encoder_out[i:i+1, :encoder_out_lens[i]]
-        else:
+            # fmt: on
-            raise ValueError(
+            if params.decoding_method == "greedy_search":
-                f"Unsupported decoding method: {params.decoding_method}"
+                hyp = greedy_search(
-            )
+                    model=model,
-        hyps.append(sp.decode(hyp).split())
+                    encoder_out=encoder_out_i,
+                    max_sym_per_frame=params.max_sym_per_frame,
+                )
+            elif params.decoding_method == "beam_search":
+                hyp = beam_search(
+                    model=model,
+                    encoder_out=encoder_out_i,
+                    beam=params.beam_size,
+                )
+            elif params.decoding_method == "modified_beam_search":
+                hyp = modified_beam_search(
+                    model=model,
+                    encoder_out=encoder_out_i,
+                    beam=params.beam_size,
+                )
+            else:
+                raise ValueError(
+                    f"Unsupported decoding method: {params.decoding_method}"
+                )
+            hyps.append(sp.decode(hyp).split())
 
     if params.decoding_method == "greedy_search":
         return {"greedy_search": hyps}
+    elif params.decoding_method == "fast_beam_search":
+        return {
+            (
+                f"beam_{params.beam}_"
+                f"max_contexts_{params.max_contexts}_"
+                f"max_states_{params.max_states}"
+            ): hyps
+        }
     else:
-        return {f"beam_{params.beam_size}": hyps}
+        return {f"beam_size_{params.beam_size}": hyps}
 
 
 def decode_dataset(
@@ -221,6 +293,7 @@ def decode_dataset(
     params: AttributeDict,
     model: nn.Module,
     sp: spm.SentencePieceProcessor,
+    decoding_graph: k2.Fsa,
 ) -> Dict[str, List[Tuple[List[str], List[str]]]]:
     """Decode dataset.
 
@@ -233,6 +306,9 @@ def decode_dataset(
         The neural model.
       sp:
         The BPE model.
+      decoding_graph:
+        The decoding graph. Can be either a `k2.TrivialGraph` or HLG, Used
+        only when --decoding_method is fast_beam_search.
     Returns:
       Return a dict, whose key may be "greedy_search" if greedy search
       is used, or it may be "beam_7" if beam size of 7 is used.
@@ -260,6 +336,7 @@ def decode_dataset(
             params=params,
             model=model,
             sp=sp,
+            decoding_graph=decoding_graph,
             batch=batch,
         )
 
@@ -340,12 +417,17 @@ def main():
     assert params.decoding_method in (
         "greedy_search",
         "beam_search",
+        "fast_beam_search",
         "modified_beam_search",
     )
     params.res_dir = params.exp_dir / params.decoding_method
 
     params.suffix = f"epoch-{params.epoch}-avg-{params.avg}"
-    if "beam_search" in params.decoding_method:
+    if "fast_beam_search" in params.decoding_method:
+        params.suffix += f"-beam-{params.beam}"
+        params.suffix += f"-max-contexts-{params.max_contexts}"
+        params.suffix += f"-max-states-{params.max_states}"
+    elif "beam_search" in params.decoding_method:
         params.suffix += f"-beam-{params.beam_size}"
     else:
         params.suffix += f"-context-{params.context_size}"
@@ -388,6 +470,11 @@ def main():
     model.eval()
     model.device = device
 
+    if params.decoding_method == "fast_beam_search":
+        decoding_graph = k2.TrivialGraph(params.vocab_size - 1, device=device)
+    else:
+        decoding_graph = None
+
     num_param = sum([p.numel() for p in model.parameters()])
     logging.info(f"Number of model parameters: {num_param}")
 
@@ -408,6 +495,7 @@ def main():
             params=params,
             model=model,
             sp=sp,
+            decoding_graph=decoding_graph,
         )
 
         save_results(

egs/librispeech/ASR/pruned_transducer_stateless/decoder.py

@@ -61,6 +61,7 @@ class Decoder(nn.Module):
 
         assert context_size >= 1, context_size
         self.context_size = context_size
+        self.vocab_size = vocab_size
         if context_size > 1:
             self.conv = nn.Conv1d(
                 in_channels=embedding_dim,