fix ctc alignment

deepspeech/exps/u2/model.py

@@ -39,6 +39,7 @@ from deepspeech.utils import error_rate
 from deepspeech.utils import layer_tools
 from deepspeech.utils import mp_tools
 from deepspeech.utils import text_grid
+from deepspeech.utils import utility
 from deepspeech.utils.log import Log
 
 logger = Log(__name__).getlog()
@@ -280,7 +281,15 @@ class U2Trainer(Trainer):
             shuffle=False,
             drop_last=False,
             collate_fn=SpeechCollator.from_config(config))
-        logger.info("Setup train/valid/test Dataloader!")
+        # return text token id
+        config.collator.keep_transcription_text = False
+        self.align_loader = DataLoader(
+            test_dataset,
+            batch_size=config.decoding.batch_size,
+            shuffle=False,
+            drop_last=False,
+            collate_fn=SpeechCollator.from_config(config))
+        logger.info("Setup train/valid/test/align Dataloader!")
 
     def setup_model(self):
         config = self.config
@@ -507,16 +516,17 @@ class U2Tester(U2Trainer):
             sys.exit(1)
 
         # xxx.align
-        assert self.args.result_file
+        assert self.args.result_file and self.args.result_file.endswith(
+            '.align')
 
         self.model.eval()
-        logger.info(f"Align Total Examples: {len(self.test_loader.dataset)}")
+        logger.info(f"Align Total Examples: {len(self.align_loader.dataset)}")
 
-        stride_ms = self.test_loader.collate_fn.stride_ms
-        token_dict = self.test_loader.collate_fn.vocab_list
+        stride_ms = self.align_loader.collate_fn.stride_ms
+        token_dict = self.align_loader.collate_fn.vocab_list
         with open(self.args.result_file, 'w') as fout:
             # one example in batch
-            for i, batch in enumerate(self.test_loader):
+            for i, batch in enumerate(self.align_loader):
                 key, feat, feats_length, target, target_length = batch
 
                 # 1. Encoder
@@ -527,36 +537,36 @@ class U2Tester(U2Trainer):
                     encoder_out)  # (1, maxlen, vocab_size)
 
                 # 2. alignment
-                # print(ctc_probs.size(1))
                 ctc_probs = ctc_probs.squeeze(0)
                 target = target.squeeze(0)
                 alignment = ctc_utils.forced_align(ctc_probs, target)
-                print(kye[0], alignment)
+                logger.info("align ids", key[0], alignment)
                 fout.write('{} {}\n'.format(key[0], alignment))
 
                 # 3. gen praat
                 # segment alignment
                 align_segs = text_grid.segment_alignment(alignment)
-                print(kye[0], align_segs)
+                logger.info("align tokens", key[0], align_segs)
                 # IntervalTier, List["start end token\n"]
-                subsample = get_subsample(self.config)
+                subsample = utility.get_subsample(self.config)
                 tierformat = text_grid.align_to_tierformat(
                     align_segs, subsample, token_dict)
                 # write tier
-                tier_path = os.path.join(
-                    os.path.dirname(args.result_file), key[0] + ".tier")
+                align_output_path = os.path.join(
+                    os.path.dirname(self.args.result_file), "align")
+                tier_path = os.path.join(align_output_path, key[0] + ".tier")
                 with open(tier_path, 'w') as f:
                     f.writelines(tierformat)
                 # write textgrid
-                textgrid_path = s.path.join(
-                    os.path.dirname(args.result_file), key[0] + ".TextGrid")
+                textgrid_path = os.path.join(align_output_path,
+                                             key[0] + ".TextGrid")
                 second_per_frame = 1. / (1000. /
                                          stride_ms)  # 25ms window, 10ms stride
                 second_per_example = (
                     len(alignment) + 1) * subsample * second_per_frame
                 text_grid.generate_textgrid(
                     maxtime=second_per_example,
-                    lines=tierformat,
+                    intervals=tierformat,
                     output=textgrid_path)
 
     def run_align(self):

deepspeech/utils/ctc_utils.py

@@ -86,13 +86,15 @@ def forced_align(ctc_probs: paddle.Tensor, y: paddle.Tensor,
     log_alpha = paddle.zeros(
         (ctc_probs.size(0), len(y_insert_blank)))  #(T, 2L+1)
     log_alpha = log_alpha - float('inf')  # log of zero
+    # TODO(Hui Zhang): zeros not support paddle.int16
     state_path = (paddle.zeros(
-        (ctc_probs.size(0), len(y_insert_blank)), dtype=paddle.int16) - 1
+        (ctc_probs.size(0), len(y_insert_blank)), dtype=paddle.int32) - 1
                   )  # state path, Tuple((T, 2L+1))
 
     # init start state
-    log_alpha[0, 0] = ctc_probs[0][y_insert_blank[0]]  # State-b, Sb
-    log_alpha[0, 1] = ctc_probs[0][y_insert_blank[1]]  # State-nb, Snb
+    # TODO(Hui Zhang): VarBase.__getitem__() not support np.int64
+    log_alpha[0, 0] = ctc_probs[0][int(y_insert_blank[0])]  # State-b, Sb
+    log_alpha[0, 1] = ctc_probs[0][int(y_insert_blank[1])]  # State-nb, Snb
 
     for t in range(1, ctc_probs.size(0)):  # T
         for s in range(len(y_insert_blank)):  # 2L+1
@@ -108,11 +110,13 @@ def forced_align(ctc_probs: paddle.Tensor, y: paddle.Tensor,
                     log_alpha[t - 1, s - 2],
                 ])
                 prev_state = [s, s - 1, s - 2]
-            log_alpha[t, s] = paddle.max(candidates) + ctc_probs[t][
-                y_insert_blank[s]]
+            # TODO(Hui Zhang): VarBase.__getitem__() not support np.int64
+            log_alpha[t, s] = paddle.max(candidates) + ctc_probs[t][int(
+                y_insert_blank[s])]
             state_path[t, s] = prev_state[paddle.argmax(candidates)]
 
-    state_seq = -1 * paddle.ones((ctc_probs.size(0), 1), dtype=paddle.int16)
+    # TODO(Hui Zhang): zeros not support paddle.int16
+    state_seq = -1 * paddle.ones((ctc_probs.size(0), 1), dtype=paddle.int32)
 
     candidates = paddle.to_tensor([
         log_alpha[-1, len(y_insert_blank) - 1],  # Sb

deepspeech/utils/text_grid.py

@@ -110,7 +110,7 @@ def generate_textgrid(maxtime: float,
     """
     # Download Praat: https://www.fon.hum.uva.nl/praat/
     avg_interval = maxtime / (len(intervals) + 1)
-    print(f"average duration per {name}: {avg_interval}")
+    print(f"average second/token: {avg_interval}")
     margin = 0.0001
 
     tg = textgrid.TextGrid(maxTime=maxtime)

deepspeech/utils/utility.py

@@ -79,3 +79,22 @@ def log_add(args: List[int]) -> float:
     a_max = max(args)
     lsp = math.log(sum(math.exp(a - a_max) for a in args))
     return a_max + lsp
+
+
+def get_subsample(config):
+    """Subsample rate from config.
+
+    Args:
+        config (yacs.config.CfgNode): yaml config
+
+    Returns:
+        int: subsample rate.
+    """
+    input_layer = config["model"]["encoder_conf"]["input_layer"]
+    assert input_layer in ["conv2d", "conv2d6", "conv2d8"]
+    if input_layer == "conv2d":
+        return 4
+    elif input_layer == "conv2d6":
+        return 6
+    elif input_layer == "conv2d8":
+        return 8

examples/aishell/s1/local/align.sh

+#! /usr/bin/env bash
+
+if [ $# != 2 ];then
+    echo "usage: ${0} config_path ckpt_path_prefix"
+    exit -1
+fi
+
+ngpu=$(echo $CUDA_VISIBLE_DEVICES | awk -F "," '{print NF}')
+echo "using $ngpu gpus..."
+
+device=gpu
+if [ ngpu == 0 ];then
+    device=cpu
+fi
+config_path=$1
+ckpt_prefix=$2
+
+ckpt_name=$(basename ${ckpt_prefxi})
+
+mkdir -p exp
+
+
+
+batch_size=1
+output_dir=${ckpt_prefix}
+mkdir -p ${output_dir}
+
+# align dump in `result_file`
+# .tier, .TextGrid dump in `dir of result_file`
+python3 -u ${BIN_DIR}/alignment.py \
+--device ${device} \
+--nproc 1 \
+--config ${config_path} \
+--result_file ${output_dir}/${type}.align \
+--checkpoint_path ${ckpt_prefix} \
+--opts decoding.batch_size ${batch_size}
+
+if [ $? -ne 0 ]; then
+    echo "Failed in ctc alignment!"
+    exit 1
+fi
+
+exit 0

tools/Makefile

@@ -19,7 +19,7 @@ kenlm.done:
 	apt-get install -y gcc-5 g++-5 && update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-5 50  && update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-5 50
 	test -d kenlm || wget -O - https://kheafield.com/code/kenlm.tar.gz | tar xz
 	mkdir -p kenlm/build && cd kenlm/build && cmake .. && make -j4 && make install
-	cd kenlm && python setup.py install
+	source venv/bin/activate; cd kenlm && python setup.py install
 	touch kenlm.done
 
 sox.done: