Skip to content

Speakr is a personal, self-hosted web application designed for transcribing audio recordings

Python

Fix speaker snippet extraction failing for video files with AAC audio

Murtaza Nasirb286b8bd3382
diff --git a/src/api/speakers.py b/src/api/speakers.py
index ed23511f..2151edf6 100644
--- a/src/api/speakers.py
+++ b/src/api/speakers.py
@@ -16,6 +16,7 @@ from src.database import db
from src.models import *
from src.utils import *
from src.utils.ffmpeg_utils import extract_audio_segment, FFmpegError, FFmpegNotFoundError
+from src.utils.ffprobe import get_codec_info, FFProbeError
from src.services.speaker_embedding_matcher import find_matching_speakers
from src.services.speaker_snippets import get_speaker_snippets, get_speaker_recordings_with_snippets
from src.services.speaker_merge import merge_speakers, preview_merge, can_merge_speakers
@@ -448,8 +449,30 @@ def get_snippet_audio(recording_id):
if not recording.audio_path or not os.path.exists(recording.audio_path):
return jsonify({'error': 'Audio file not found'}), 404
+ # Detect audio codec to pick the right output container for stream copy
+ codec_to_container = {
+ 'mp3': ('.mp3', 'audio/mpeg'),
+ 'aac': ('.m4a', 'audio/mp4'),
+ 'opus': ('.ogg', 'audio/ogg'),
+ 'vorbis': ('.ogg', 'audio/ogg'),
+ 'flac': ('.flac', 'audio/flac'),
+ 'pcm_s16le': ('.wav', 'audio/wav'),
+ 'pcm_s24le': ('.wav', 'audio/wav'),
+ 'pcm_s32le': ('.wav', 'audio/wav'),
+ 'pcm_f32le': ('.wav', 'audio/wav'),
+ }
+ snippet_ext = '.mp3'
+ snippet_mime = 'audio/mpeg'
+ try:
+ codec_info = get_codec_info(recording.audio_path, timeout=10)
+ audio_codec = codec_info.get('audio_codec')
+ if audio_codec and audio_codec in codec_to_container:
+ snippet_ext, snippet_mime = codec_to_container[audio_codec]
+ except FFProbeError:
+ pass # Fall back to mp3
+
# Create temporary file for the snippet
- with tempfile.NamedTemporaryFile(delete=False, suffix='.mp3') as tmp_file:
+ with tempfile.NamedTemporaryFile(delete=False, suffix=snippet_ext) as tmp_file:
output_path = tmp_file.name
try:
@@ -464,9 +487,9 @@ def get_snippet_audio(recording_id):
# Send the file
response = send_file(
output_path,
- mimetype='audio/mpeg',
+ mimetype=snippet_mime,
as_attachment=False,
- download_name=f'snippet_{recording_id}_{start_time:.1f}s.mp3'
+ download_name=f'snippet_{recording_id}_{start_time:.1f}s{snippet_ext}'
)
# Clean up temporary file after sending
diff --git a/src/utils/ffmpeg_utils.py b/src/utils/ffmpeg_utils.py
index 50afba84..50725280 100644
--- a/src/utils/ffmpeg_utils.py
+++ b/src/utils/ffmpeg_utils.py
@@ -367,7 +367,8 @@ def extract_audio_segment(
'-i', input_path,
'-ss', str(start_time),
'-t', str(duration),
- '-c', 'copy', # Copy codec (no re-encoding)
+ '-vn', # Drop video streams (audio segment only)
+ '-c:a', 'copy', # Copy audio codec (no re-encoding)
'-y',
output_path
]