diff --git a/medkit/audio/transcription/sb_transcriber.py b/medkit/audio/transcription/sb_transcriber.py
index 647d4cab..ea666e4c 100644
--- a/medkit/audio/transcription/sb_transcriber.py
+++ b/medkit/audio/transcription/sb_transcriber.py
@@ -89,7 +89,9 @@ def __init__(
         self.batch_size = batch_size
         self._torch_device = "cpu" if self.device < 0 else f"cuda:{self.device}"
 
-        asr_class = speechbrain.pretrained.EncoderDecoderASR if needs_decoder else speechbrain.pretrained.EncoderASR
+        asr_class = (
+            speechbrain.inference.ASR.EncoderDecoderASR if needs_decoder else speechbrain.inference.ASR.EncoderASR
+        )
 
         self._asr = asr_class.from_hparams(source=model, savedir=cache_dir, run_opts={"device": self._torch_device})
 
diff --git a/pyproject.toml b/pyproject.toml
index 13c885ec..79ee606c 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -90,7 +90,7 @@ metrics-text-classification = [
   "scikit-learn>=1.3.2",
 ]
 metrics-transcription = [
-  "speechbrain>=0.5",
+  "speechbrain>=1.0",
 ]
 nlstruct = [
   "huggingface-hub",
diff --git a/tests/unit/audio/transcription/test_sb_transcriber.py b/tests/unit/audio/transcription/test_sb_transcriber.py
index 00590658..8bac4b10 100644
--- a/tests/unit/audio/transcription/test_sb_transcriber.py
+++ b/tests/unit/audio/transcription/test_sb_transcriber.py
@@ -56,8 +56,8 @@ def __init__(self):
 
 @pytest.fixture(scope="module", autouse=True)
 def _mocked_asr(module_mocker):
-    module_mocker.patch("speechbrain.pretrained.EncoderASR", _MockSpeechbrainASR)
-    module_mocker.patch("speechbrain.pretrained.EncoderDecoderASR", _MockSpeechbrainASR)
+    module_mocker.patch("speechbrain.inference.ASR.EncoderASR", _MockSpeechbrainASR)
+    module_mocker.patch("speechbrain.inference.ASR.EncoderDecoderASR", _MockSpeechbrainASR)
 
 
 def _gen_segment(nb_samples) -> Segment: