Skip to content

[FEATURE_REQUEST] Add transcribe vad params #227

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
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

[FEATURE_REQUEST] Add transcribe vad params #227

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
10e7cdc
add transcribe vad params
jspython Nov 25, 2024
1aae839
reset params default values
jspython Nov 25, 2024
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
26 changes: 25 additions & 1 deletion whisper_timestamped/transcribe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -117,6 +117,10 @@ def transcribe_timestamped(
suppress_tokens="-1",
sample_len=None,
verbose=False,
avoid_empty_speech=True,
vad_min_speech_duration=0.1,
vad_min_silence_duration=1,
vad_dilatation=0.5,
):
"""
Transcribe an audio file using Whisper
Expand Down Expand Up @@ -214,6 +218,18 @@ def transcribe_timestamped(
Whether to display the text being decoded to the console. If True, displays all the details,
If False, displays minimal details. If None, does not display anything

avoid_empty_speech: bool
Whether to avoid empty speech segments (i.e. segments with no speech detected).

vad_min_speech_duration: float
Minimum duration of a speech segment, in seconds. If a speech segment is shorter than this, it will be removed.

vad_min_silence_duration: float
Minimum duration of a silence segment, in seconds. If a silence segment is shorter than this, it will be removed.

vad_dilatation: float
Dilatation factor for the speech segments. If a speech segment is shorter than this, it will be removed.

Returns
-------
A dictionary containing the resulting text ("text") and segment-level details ("segments"), and
Expand Down Expand Up @@ -293,7 +309,15 @@ def transcribe_timestamped(

if vad is not None:
audio = get_audio_tensor(audio)
audio, vad_segments, convert_timestamps = remove_non_speech(audio, method=vad, sample_rate=SAMPLE_RATE, plot=plot_word_alignment, avoid_empty_speech=True)
audio, vad_segments, convert_timestamps = remove_non_speech(audio,
method=vad,
sample_rate=SAMPLE_RATE,
plot=plot_word_alignment,
avoid_empty_speech=avoid_empty_speech,
min_speech_duration=vad_min_speech_duration,
min_silence_duration=vad_min_silence_duration,
dilatation=vad_dilatation,
)
else:
vad_segments = None

Expand Down