Skip to content

fix: Gracefully handle missing speech profiles bucket #6223

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
neooriginal wants to merge 2 commits into
base: main
Choose a base branch
from
+67 −15
Open

fix: Gracefully handle missing speech profiles bucket #6223

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5e83c95
Gracefully handle missing speech profiles bucket
neooriginal Apr 1, 2026
bff392d
Merge branch 'BasedHardware:main' into main
neooriginal Apr 1, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
82 changes: 67 additions & 15 deletions backend/utils/other/storage.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@
else:
storage_client = storage.Client()

speech_profiles_bucket = os.getenv('BUCKET_SPEECH_PROFILES')
speech_profiles_bucket = (os.getenv('BUCKET_SPEECH_PROFILES') or '').strip() or None
postprocessing_audio_bucket = os.getenv('BUCKET_POSTPROCESSING')
memories_recordings_bucket = os.getenv('BUCKET_MEMORIES_RECORDINGS')
private_cloud_sync_bucket = os.getenv('BUCKET_PRIVATE_CLOUD_SYNC', 'omi-private-cloud-sync')
Expand All @@ -48,20 +48,41 @@
chat_files_bucket = os.getenv('BUCKET_CHAT_FILES')
desktop_updates_bucket = os.getenv('BUCKET_DESKTOP_UPDATES')

_did_warn_missing_speech_profiles_bucket = False


def _get_speech_profiles_bucket(required: bool = False):
global _did_warn_missing_speech_profiles_bucket

if speech_profiles_bucket:
return storage_client.bucket(speech_profiles_bucket)

if not _did_warn_missing_speech_profiles_bucket:
logger.warning('BUCKET_SPEECH_PROFILES is not configured; speech profile storage is disabled')
_did_warn_missing_speech_profiles_bucket = True

if required:
raise RuntimeError('BUCKET_SPEECH_PROFILES is not configured')
Comment thread
neooriginal marked this conversation as resolved.

Comment thread
neooriginal marked this conversation as resolved.
return None


# *******************************************
# ************* SPEECH PROFILE **************
# *******************************************
def upload_profile_audio(file_path: str, uid: str):
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket(required=True)
path = f'{uid}/speech_profile.wav'
blob = bucket.blob(path)
blob.upload_from_filename(file_path)
return f'https://storage.googleapis.com/{speech_profiles_bucket}/{path}'


def get_user_has_speech_profile(uid: str, max_age_days: int = None) -> bool:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket()
if bucket is None:
return False

blob = bucket.blob(f'{uid}/speech_profile.wav')
if not blob.exists():
return False
Expand All @@ -78,7 +99,10 @@ def get_user_has_speech_profile(uid: str, max_age_days: int = None) -> bool:


def get_profile_audio_if_exists(uid: str, download: bool = True) -> str:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket()
if bucket is None:
return None

Comment thread
neooriginal marked this conversation as resolved.
path = f'{uid}/speech_profile.wav'
blob = bucket.blob(path)
if blob.exists():
Expand All @@ -92,22 +116,28 @@ def get_profile_audio_if_exists(uid: str, download: bool = True) -> str:


def upload_additional_profile_audio(file_path: str, uid: str) -> None:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket(required=True)
path = f'{uid}/additional_profile_recordings/{file_path.split("/")[-1]}'
blob = bucket.blob(path)
blob.upload_from_filename(file_path)


def delete_additional_profile_audio(uid: str, file_name: str) -> None:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket()
if bucket is None:
return

blob = bucket.blob(f'{uid}/additional_profile_recordings/{file_name}')
if blob.exists():
logger.info(f'delete_additional_profile_audio deleting {file_name}')
blob.delete()


def get_additional_profile_recordings(uid: str, download: bool = False) -> List[str]:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket()
if bucket is None:
return []

blobs = bucket.list_blobs(prefix=f'{uid}/additional_profile_recordings/')
if download:
paths = []
Expand All @@ -126,21 +156,27 @@ def get_additional_profile_recordings(uid: str, download: bool = False) -> List[


def upload_user_person_speech_sample(file_path: str, uid: str, person_id: str) -> None:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket(required=True)
path = f'{uid}/people_profiles/{person_id}/{file_path.split("/")[-1]}'
blob = bucket.blob(path)
blob.upload_from_filename(file_path)


def delete_user_person_speech_sample(uid: str, person_id: str, file_name: str) -> None:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket()
if bucket is None:
return

blob = bucket.blob(f'{uid}/people_profiles/{person_id}/{file_name}')
if blob.exists():
blob.delete()


def delete_speech_sample_for_people(uid: str, file_name: str) -> None:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket()
if bucket is None:
return

blobs = bucket.list_blobs(prefix=f'{uid}/people_profiles/')
for blob in blobs:
if file_name in blob.name:
Expand All @@ -149,7 +185,10 @@ def delete_speech_sample_for_people(uid: str, file_name: str) -> None:


def delete_user_person_speech_samples(uid: str, person_id: str) -> None:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket()
if bucket is None:
return

blobs = bucket.list_blobs(prefix=f'{uid}/people_profiles/{person_id}/')
for blob in blobs:
blob.delete()
Expand All @@ -171,7 +210,7 @@ def upload_person_speech_sample_from_bytes(
wav_file.setframerate(sample_rate)
wav_file.writeframes(audio_bytes)

bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket(required=True)
filename = f"{uuid_module.uuid4()}.wav"
path = f'{uid}/people_profiles/{person_id}/{filename}'
blob = bucket.blob(path)
Expand All @@ -181,13 +220,19 @@ def upload_person_speech_sample_from_bytes(


def get_user_people_ids(uid: str) -> List[str]:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket()
if bucket is None:
return []

blobs = bucket.list_blobs(prefix=f'{uid}/people_profiles/')
return [blob.name.split("/")[-2] for blob in blobs]


def get_user_person_speech_samples(uid: str, person_id: str, download: bool = False) -> List[str]:
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket()
if bucket is None:
return []

blobs = bucket.list_blobs(prefix=f'{uid}/people_profiles/{person_id}/')
if download:
paths = []
Expand All @@ -213,7 +258,10 @@ def get_speech_sample_signed_urls(paths: List[str]) -> List[str]:
"""
if not paths:
return []
bucket = storage_client.bucket(speech_profiles_bucket)
bucket = _get_speech_profiles_bucket()
if bucket is None:
return []

signed_urls = []
for path in paths:
blob = bucket.blob(path)
Expand Down Expand Up @@ -1087,6 +1135,8 @@ def download_speech_profile_bytes(path: str) -> bytes:
Raises:
NotFound: If the sample doesn't exist
"""
if not speech_profiles_bucket:
raise BlobNotFound('Speech profile storage is not configured')
Comment thread
neooriginal marked this conversation as resolved.
Comment thread
neooriginal marked this conversation as resolved.
return download_blob_bytes(speech_profiles_bucket, path)


Expand All @@ -1100,6 +1150,8 @@ def delete_speech_profile_blob(path: str) -> bool:
Returns:
True if deleted, False if not found
"""
if not speech_profiles_bucket:
Comment thread
neooriginal marked this conversation as resolved.
return False
return delete_blob(speech_profiles_bucket, path)


Expand Down
Loading