Mod: Update UI to store microphone input in microphone_conette_record.wav file, raises an error when the audio is too short or too long, update main description and show other candidates in outputs.
Browse files- .gitignore +1 -1
- app.py +63 -26
.gitignore
CHANGED
@@ -1 +1 @@
|
|
1 |
-
|
|
|
1 |
+
microphone_conette_record.wav
|
app.py
CHANGED
@@ -2,12 +2,14 @@
|
|
2 |
# -*- coding: utf-8 -*-
|
3 |
|
4 |
from tempfile import NamedTemporaryFile, _TemporaryFileWrapper
|
5 |
-
from typing import Any, Optional
|
6 |
|
7 |
import streamlit as st
|
|
|
8 |
|
9 |
from st_audiorec import st_audiorec
|
10 |
from streamlit.runtime.uploaded_file_manager import UploadedFile
|
|
|
11 |
|
12 |
from conette import CoNeTTEModel, conette
|
13 |
from conette.utils.collections import dict_list_to_list_dict
|
@@ -17,9 +19,11 @@ ALLOW_REP_MODES = ("stopwords", "all", "none")
|
|
17 |
MAX_BEAM_SIZE = 20
|
18 |
MAX_PRED_SIZE = 30
|
19 |
MAX_BATCH_SIZE = 32
|
20 |
-
RECORD_AUDIO_FNAME = "
|
21 |
DEFAULT_THRESHOLD = 0.3
|
22 |
THRESHOLD_PRECISION = 100
|
|
|
|
|
23 |
|
24 |
|
25 |
@st.cache_resource
|
@@ -49,20 +53,34 @@ def get_results(
|
|
49 |
model: CoNeTTEModel,
|
50 |
audio_files: dict[str, bytes],
|
51 |
generate_kwds: dict[str, Any],
|
52 |
-
) -> dict[str, dict[str, Any]]:
|
53 |
# Get audio to be processed
|
54 |
-
audio_to_predict: dict[str, bytes] = {}
|
55 |
for audio_fname, audio in audio_files.items():
|
56 |
result_hash = get_result_hash(audio_fname, generate_kwds)
|
57 |
if result_hash not in st.session_state or audio_fname == RECORD_AUDIO_FNAME:
|
58 |
-
audio_to_predict[result_hash] = audio
|
59 |
|
60 |
# Save audio to be processed
|
61 |
tmp_files: dict[str, _TemporaryFileWrapper] = {}
|
62 |
-
for result_hash, audio in audio_to_predict.items():
|
63 |
-
tmp_file = NamedTemporaryFile()
|
64 |
tmp_file.write(audio)
|
65 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
66 |
|
67 |
# Generate predictions and store them in session state
|
68 |
for start in range(0, len(tmp_files), MAX_BATCH_SIZE):
|
@@ -74,8 +92,6 @@ def get_results(
|
|
74 |
tmp_paths_j,
|
75 |
**generate_kwds,
|
76 |
)
|
77 |
-
for tmp_file in tmp_files_j:
|
78 |
-
tmp_file.close()
|
79 |
outputs_lst = dict_list_to_list_dict(outputs_j) # type: ignore
|
80 |
for result_hash, output_i in zip(result_hashes_j, outputs_lst):
|
81 |
st.session_state[result_hash] = output_i
|
@@ -90,46 +106,67 @@ def get_results(
|
|
90 |
return outputs
|
91 |
|
92 |
|
93 |
-
def show_results(outputs: dict[str, dict[str, Any]]) -> None:
|
94 |
st.divider()
|
95 |
|
96 |
for audio_fname, output in outputs.items():
|
97 |
-
|
98 |
-
|
99 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
100 |
|
101 |
cand = format_candidate(cand)
|
102 |
-
tags = format_tags(tags)
|
103 |
prob = lprobs.exp().tolist()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
104 |
|
105 |
if audio_fname == RECORD_AUDIO_FNAME:
|
106 |
header = "##### Result for microphone input:"
|
107 |
else:
|
108 |
header = f'##### Result for "{audio_fname}"'
|
109 |
|
110 |
-
content =
|
111 |
-
|
112 |
-
|
113 |
-
|
114 |
-
|
115 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
116 |
st.divider()
|
117 |
|
118 |
|
119 |
def main() -> None:
|
120 |
-
st.header("Describe audio content with CoNeTTE")
|
121 |
-
|
122 |
model = load_conette(model_kwds=dict(device="cpu"))
|
123 |
|
124 |
-
|
125 |
-
|
126 |
-
|
|
|
127 |
|
128 |
record_data = st_audiorec()
|
129 |
audio_files: Optional[list[UploadedFile]] = st.file_uploader(
|
130 |
"**Or upload audio files here:**",
|
131 |
type=["wav", "flac", "mp3", "ogg", "avi"],
|
132 |
accept_multiple_files=True,
|
|
|
133 |
)
|
134 |
|
135 |
with st.expander("Model hyperparameters"):
|
|
|
2 |
# -*- coding: utf-8 -*-
|
3 |
|
4 |
from tempfile import NamedTemporaryFile, _TemporaryFileWrapper
|
5 |
+
from typing import Any, Optional, Union
|
6 |
|
7 |
import streamlit as st
|
8 |
+
import torchaudio
|
9 |
|
10 |
from st_audiorec import st_audiorec
|
11 |
from streamlit.runtime.uploaded_file_manager import UploadedFile
|
12 |
+
from torch import Tensor
|
13 |
|
14 |
from conette import CoNeTTEModel, conette
|
15 |
from conette.utils.collections import dict_list_to_list_dict
|
|
|
19 |
MAX_BEAM_SIZE = 20
|
20 |
MAX_PRED_SIZE = 30
|
21 |
MAX_BATCH_SIZE = 32
|
22 |
+
RECORD_AUDIO_FNAME = "microphone_conette_record.wav"
|
23 |
DEFAULT_THRESHOLD = 0.3
|
24 |
THRESHOLD_PRECISION = 100
|
25 |
+
MIN_AUDIO_DURATION_SEC = 0.3
|
26 |
+
MAX_AUDIO_DURATION_SEC = 60
|
27 |
|
28 |
|
29 |
@st.cache_resource
|
|
|
53 |
model: CoNeTTEModel,
|
54 |
audio_files: dict[str, bytes],
|
55 |
generate_kwds: dict[str, Any],
|
56 |
+
) -> dict[str, Union[dict[str, Any], str]]:
|
57 |
# Get audio to be processed
|
58 |
+
audio_to_predict: dict[str, tuple[str, bytes]] = {}
|
59 |
for audio_fname, audio in audio_files.items():
|
60 |
result_hash = get_result_hash(audio_fname, generate_kwds)
|
61 |
if result_hash not in st.session_state or audio_fname == RECORD_AUDIO_FNAME:
|
62 |
+
audio_to_predict[result_hash] = (audio_fname, audio)
|
63 |
|
64 |
# Save audio to be processed
|
65 |
tmp_files: dict[str, _TemporaryFileWrapper] = {}
|
66 |
+
for result_hash, (audio_fname, audio) in audio_to_predict.items():
|
67 |
+
tmp_file = NamedTemporaryFile(delete=False)
|
68 |
tmp_file.write(audio)
|
69 |
+
tmp_file.close()
|
70 |
+
|
71 |
+
metadata = torchaudio.info(tmp_file.name) # type: ignore
|
72 |
+
duration = metadata.num_frames / metadata.sample_rate
|
73 |
+
|
74 |
+
if MIN_AUDIO_DURATION_SEC > duration:
|
75 |
+
error_msg = f"Audio file is too short. (found {duration:.2f}s but the model expect audio in range [{MIN_AUDIO_DURATION_SEC}, {MAX_AUDIO_DURATION_SEC}])"
|
76 |
+
st.session_state[result_hash] = error_msg
|
77 |
+
|
78 |
+
elif duration > MAX_AUDIO_DURATION_SEC:
|
79 |
+
error_msg = f"Audio file is too long. (found {duration:.2f}s but the model expect audio in range [{MIN_AUDIO_DURATION_SEC}, {MAX_AUDIO_DURATION_SEC}])"
|
80 |
+
st.session_state[result_hash] = error_msg
|
81 |
+
|
82 |
+
else:
|
83 |
+
tmp_files[result_hash] = tmp_file
|
84 |
|
85 |
# Generate predictions and store them in session state
|
86 |
for start in range(0, len(tmp_files), MAX_BATCH_SIZE):
|
|
|
92 |
tmp_paths_j,
|
93 |
**generate_kwds,
|
94 |
)
|
|
|
|
|
95 |
outputs_lst = dict_list_to_list_dict(outputs_j) # type: ignore
|
96 |
for result_hash, output_i in zip(result_hashes_j, outputs_lst):
|
97 |
st.session_state[result_hash] = output_i
|
|
|
106 |
return outputs
|
107 |
|
108 |
|
109 |
+
def show_results(outputs: dict[str, Union[dict[str, Any], str]]) -> None:
|
110 |
st.divider()
|
111 |
|
112 |
for audio_fname, output in outputs.items():
|
113 |
+
if isinstance(output, str):
|
114 |
+
st.error(output)
|
115 |
+
st.divider()
|
116 |
+
continue
|
117 |
+
|
118 |
+
cand: str = output["cands"]
|
119 |
+
lprobs: Tensor = output["lprobs"]
|
120 |
+
tags_lst = output.get("tags")
|
121 |
+
mult_cands: list[str] = output["mult_cands"]
|
122 |
+
mult_lprobs: Tensor = output["mult_lprobs"]
|
123 |
|
124 |
cand = format_candidate(cand)
|
|
|
125 |
prob = lprobs.exp().tolist()
|
126 |
+
tags = format_tags(tags_lst)
|
127 |
+
mult_cands = [format_candidate(cand_i) for cand_i in mult_cands]
|
128 |
+
mult_probs = mult_lprobs.exp()
|
129 |
+
|
130 |
+
indexes = mult_probs.argsort(descending=True)[1:]
|
131 |
+
mult_probs = mult_probs[indexes].tolist()
|
132 |
+
mult_cands = [mult_cands[idx] for idx in indexes]
|
133 |
|
134 |
if audio_fname == RECORD_AUDIO_FNAME:
|
135 |
header = "##### Result for microphone input:"
|
136 |
else:
|
137 |
header = f'##### Result for "{audio_fname}"'
|
138 |
|
139 |
+
content = [
|
140 |
+
header,
|
141 |
+
f'- **Description:** "{cand}" ({prob*100:.1f}%)',
|
142 |
+
f"- **Tags:** {tags}",
|
143 |
+
]
|
144 |
+
if len(mult_cands) > 0:
|
145 |
+
msg = f"- **Other descriptions:**"
|
146 |
+
content.append(msg)
|
147 |
+
|
148 |
+
for cand_i, prob_i in zip(mult_cands, mult_probs):
|
149 |
+
msg = f' - "{cand_i}" ({prob_i*100:.1f}%)'
|
150 |
+
content.append(msg)
|
151 |
+
|
152 |
+
st.success("\n".join(content))
|
153 |
st.divider()
|
154 |
|
155 |
|
156 |
def main() -> None:
|
|
|
|
|
157 |
model = load_conette(model_kwds=dict(device="cpu"))
|
158 |
|
159 |
+
st.header("Describe audio content with CoNeTTE")
|
160 |
+
st.markdown(
|
161 |
+
"This interface allows you to generate a short description of the sound events of any recording. You can try it from your microphone or upload a file below."
|
162 |
+
)
|
163 |
|
164 |
record_data = st_audiorec()
|
165 |
audio_files: Optional[list[UploadedFile]] = st.file_uploader(
|
166 |
"**Or upload audio files here:**",
|
167 |
type=["wav", "flac", "mp3", "ogg", "avi"],
|
168 |
accept_multiple_files=True,
|
169 |
+
help="Recommanded audio: lasting from **1 to 30s**, sampled at **32 kHz** minimum.",
|
170 |
)
|
171 |
|
172 |
with st.expander("Model hyperparameters"):
|