From 4a98772d5ed1c673fe1e686ebddb943b5e9055da Mon Sep 17 00:00:00 2001
From: Kevin Lenzo <kevinlenzo@gmail.com>
Date: Sat, 15 Aug 2026 11:19:51 -0400
Subject: [PATCH] Add a stream_reset option to pocketsphinx_batch

With noise removal enabled, batch results depend on control-file
order: one shared decoder processes every entry, and the
spectral-subtraction noise estimate persists across entries because
ps_start_utt() does not reset it. Only ps_start_stream() resets it,
and the batch tool never called it. Retention is correct when a
control file addresses segments of one continuous recording, but
batch entries are routinely independent recordings, and the tool
offered no way to choose. The same audio file listed twice in one
control file can report two different scores in a single run.

The new stream_reset option (default no, preserving current
behavior) calls ps_start_stream() before each control-file entry
when noise removal is effectively enabled.

test-batch-stream-reset.sh decodes a control file that lists the
same recording twice, with other audio between, under
-stream_reset yes, and asserts the repeated entries report
identical hypothesis and score. It fails when the reset is not
performed.
---
 doxygen/pocketsphinx_batch.1               |  3 ++
 programs/pocketsphinx_batch.c              |  7 ++++
 test/regression/CMakeLists.txt             |  1 +
 test/regression/test-batch-stream-reset.sh | 39 ++++++++++++++++++++++
 4 files changed, 50 insertions(+)
 create mode 100644 test/regression/test-batch-stream-reset.sh

diff --git a/doxygen/pocketsphinx_batch.1 b/doxygen/pocketsphinx_batch.1
index 8e660af81..ba59fa6db 100644
--- a/doxygen/pocketsphinx_batch.1
+++ b/doxygen/pocketsphinx_batch.1
@@ -369,6 +369,9 @@ Silence word transition probability
 .B \-smoothspec
 Write out cepstral-smoothed logspectral files
 .TP
+.B \-stream_reset
+Reset noise-removal statistics before each control-file entry, for control files whose entries are independent recordings
+.TP
 .B \-svspec
 specification (e.g., 24,0-11/25,12-23/26-38 or 0-12/13-25/26-38)
 .TP
diff --git a/programs/pocketsphinx_batch.c b/programs/pocketsphinx_batch.c
index 3a1ae210f..d904d3fcf 100644
--- a/programs/pocketsphinx_batch.c
+++ b/programs/pocketsphinx_batch.c
@@ -79,6 +79,10 @@ static const ps_arg_t ps_args_def[] = {
       ARG_INTEGER,
       "1",
       "Do every Nth line in the control file" },
+    { "stream_reset",
+      ARG_BOOLEAN,
+      "no",
+      "Reset noise-removal statistics before each control-file entry, for control files whose entries are independent recordings" },
     { "mllrctl",
       ARG_STRING,
       NULL,
@@ -836,6 +840,9 @@ process_ctl(ps_decoder_t *ps, cmd_ln_t *config, FILE *ctlfh)
             E_INFO("Decoding '%s'\n", uttid);
 
             /* Do actual decoding. */
+            if (ps_config_bool(config, "stream_reset")
+                && ps_config_bool(config, "remove_noise"))
+                ps_start_stream(ps);
             if(process_mllrctl_line(ps, config, mllrfile) < 0)
                 continue;
             if(process_lmnamectl_line(ps, config, lmname) < 0)
diff --git a/test/regression/CMakeLists.txt b/test/regression/CMakeLists.txt
index 4ef434d58..56ca37fa6 100644
--- a/test/regression/CMakeLists.txt
+++ b/test/regression/CMakeLists.txt
@@ -3,6 +3,7 @@ set(TESTS
   test-lm.sh
   test-lm-convert.sh
   test-main.sh
+  test-batch-stream-reset.sh
   test-main-align.sh
   test-align.sh
   test-tidigits-fsg.sh
diff --git a/test/regression/test-batch-stream-reset.sh b/test/regression/test-batch-stream-reset.sh
new file mode 100644
index 000000000..7dcfde169
--- /dev/null
+++ b/test/regression/test-batch-stream-reset.sh
@@ -0,0 +1,39 @@
+#!/bin/bash
+
+: ${CMAKE_BINARY_DIR:=$(pwd)}
+. ${CMAKE_BINARY_DIR}/test/testfuncs.sh
+
+bn=`basename $0 .sh`
+ctl=$bn.ctl
+
+echo "Test: $bn"
+
+printf 'goforward 0 -1 gof_first\nnumbers 0 -1 num\nsomething 0 -1 som\ngoforward 0 -1 gof_again\n' >$ctl
+rm -f $bn.hyp
+
+run_program pocketsphinx_batch \
+    -hmm $model/en-us/en-us \
+    -lm $model/en-us/en-us.lm.bin \
+    -dict $model/en-us/cmudict-en-us.dict \
+    -ctl $ctl \
+    -cepdir $data \
+    -cepext .raw \
+    -adcin yes \
+    -stream_reset yes \
+    -hyp $bn.hyp \
+    2>$bn.log
+
+if [ $? = 0 ]; then
+    pass "run"
+else
+    fail "run"
+fi
+
+first=`grep ' (gof_first ' $bn.hyp 2>/dev/null | sed 's/ (gof_first / (gof /'`
+again=`grep ' (gof_again ' $bn.hyp 2>/dev/null | sed 's/ (gof_again / (gof /'`
+
+if test -n "$first" && test "$first" = "$again"; then
+    pass "repeated entry"
+else
+    fail "repeated entry"
+fi