diff --git a/src/crowsetta/formats/bbox/raven.py b/src/crowsetta/formats/bbox/raven.py
index 7e28fc93..346c367b 100644
--- a/src/crowsetta/formats/bbox/raven.py
+++ b/src/crowsetta/formats/bbox/raven.py
@@ -25,8 +25,8 @@ class RavenSchema(pandera.pandas.DataFrameModel):
 
     begin_time_s: Series[float] = pandera.pandas.Field()
     end_time_s: Series[float] = pandera.pandas.Field()
-    low_freq_hz: Series[float] = pandera.pandas.Field()
-    high_freq_hz: Series[float] = pandera.pandas.Field()
+    low_freq_hz: Series[float] = pandera.pandas.Field(coerce=True)
+    high_freq_hz: Series[float] = pandera.pandas.Field(coerce=True)
     annotation: Series[pd.StringDtype] = pandera.pandas.Field(coerce=True)
 
     class Config:
diff --git a/tests/test_formats/test_bbox/test_audbbox.py b/tests/test_formats/test_bbox/test_audbbox.py
index a9a2ef9e..33675dd8 100644
--- a/tests/test_formats/test_bbox/test_audbbox.py
+++ b/tests/test_formats/test_bbox/test_audbbox.py
@@ -53,7 +53,7 @@ def test_raven_schema_bad_df(self, an_audbbox_path):
         # if we just load the csv directly, it will be invalid
         df = pd.read_csv(an_audbbox_path, sep="\t", header=None)
 
-        with pytest.raises(pandera.errors.SchemaError):
+        with pytest.raises(pandera.errors.SchemaErrors):
             crowsetta.formats.bbox.audbbox.AudBBoxSchema.validate(df)
 
 
diff --git a/tests/test_formats/test_seq/test_audseq.py b/tests/test_formats/test_seq/test_audseq.py
index d0d23dab..1ee5dafe 100644
--- a/tests/test_formats/test_seq/test_audseq.py
+++ b/tests/test_formats/test_seq/test_audseq.py
@@ -22,7 +22,7 @@ def test_audseq_schema_bad_df(self, an_audseq_path):
         audseq_df = pd.read_csv(an_audseq_path, sep="\t", header=None)
         # notice: wrong column name 'stop_time'
         audseq_df.columns = ["start_time", "stop_time", "label"]
-        with pytest.raises(pandera.errors.SchemaError):
+        with pytest.raises(pandera.errors.SchemaErrors):
             crowsetta.formats.seq.audseq.AudSeqSchema.validate(audseq_df)
 
 
diff --git a/tests/test_formats/test_seq/test_generic.py b/tests/test_formats/test_seq/test_generic.py
index d1fc70db..2b29641f 100644
--- a/tests/test_formats/test_seq/test_generic.py
+++ b/tests/test_formats/test_seq/test_generic.py
@@ -62,7 +62,7 @@ def test_missing_column_raises(self, csv_missing_fields_in_header):
     def test_invalid_column_raises(self, csv_with_invalid_fields_in_header):
         """test that invalid column name 'invalid' raises a schema error"""
         invalid_df = pd.read_csv(csv_with_invalid_fields_in_header)
-        with pytest.raises(pandera.errors.SchemaError):
+        with pytest.raises(pandera.errors.SchemaErrors):
             crowsetta.formats.seq.generic.GenericSeqSchema.validate(invalid_df)
 
     def test_onset_s_with_no_offset_s_raises(self, csv_with_onset_s_but_no_offset_s):
@@ -378,7 +378,7 @@ def test_csv2annot_missing_fields_raises(self, csv_missing_fields_in_header):
             crowsetta.formats.seq.generic.csv2annot(csv_path=csv_missing_fields_in_header)
 
     def test_csv2annot_invalid_fields_raises(self, csv_with_invalid_fields_in_header):
-        with pytest.raises(pandera.errors.SchemaError):
+        with pytest.raises(pandera.errors.SchemaErrors):
             crowsetta.formats.seq.generic.csv2annot(csv_path=csv_with_invalid_fields_in_header)
 
 
diff --git a/tests/test_formats/test_seq/test_simple.py b/tests/test_formats/test_seq/test_simple.py
index 2533b8a2..6c5ddc25 100644
--- a/tests/test_formats/test_seq/test_simple.py
+++ b/tests/test_formats/test_seq/test_simple.py
@@ -22,7 +22,7 @@ def test_simple_seq_schema_bad_df(self, a_simple_csv_path):
         simple_df = pd.read_csv(a_simple_csv_path)
         # notice: wrong column names
         simple_df.columns = ["text", "begin_sample", "end_sample"]
-        with pytest.raises(pandera.errors.SchemaError):
+        with pytest.raises(pandera.errors.SchemaErrors):
             simple_df = crowsetta.formats.seq.simple.SimpleSeqSchema.validate(simple_df)
 
 
diff --git a/tests/test_formats/test_seq/test_timit.py b/tests/test_formats/test_seq/test_timit.py
index bcc833be..cc8602f8 100644
--- a/tests/test_formats/test_seq/test_timit.py
+++ b/tests/test_formats/test_seq/test_timit.py
@@ -19,7 +19,7 @@ def test_timit_schema_bad_df(self, a_transcript_path):
         phn_df = pd.read_csv(a_transcript_path, sep=" ", header=None)
         # notice: wrong column names
         phn_df.columns = ["text", "begin_sample", "end_sample"]
-        with pytest.raises(pandera.errors.SchemaError):
+        with pytest.raises(pandera.errors.SchemaErrors):
             phn_df = crowsetta.formats.seq.timit.TimitTranscriptSchema.validate(phn_df)