Skip to content

Enable speaker diarization for ElevenLabs Scribe #106

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
louisjoecodes wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Enable speaker diarization for ElevenLabs Scribe #106

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
39 changes: 27 additions & 12 deletions api/run_eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -198,23 +198,38 @@ def transcribe_with_retry(

elif model_name.startswith("elevenlabs/"):
client = ElevenLabs(api_key=os.getenv("ELEVENLABS_API_KEY"))

api_params = {
"model_id": model_name.split("/")[1],
"language_code": "en",
"diarize": True,
"timestamps_granularity": "word",
}

if use_url:
response = requests.get(sample["row"]["audio"][0]["src"])
audio_data = BytesIO(response.content)
transcription = client.speech_to_text.convert(
file=audio_data,
model_id=model_name.split("/")[1],
language_code="eng",
tag_audio_events=True,
)
transcription = client.speech_to_text.convert(file=audio_data, **api_params)
else:
with open(audio_file_path, "rb") as audio_file:
transcription = client.speech_to_text.convert(
file=audio_file,
model_id=model_name.split("/")[1],
language_code="eng",
tag_audio_events=True,
)
transcription = client.speech_to_text.convert(file=audio_file, **api_params)

if hasattr(transcription, 'words') and transcription.words:
speaker_word_counts = {}
speaker_words = {}

for word_obj in transcription.words:
if hasattr(word_obj, 'speaker_id') and word_obj.speaker_id:
speaker_id = word_obj.speaker_id
speaker_word_counts[speaker_id] = speaker_word_counts.get(speaker_id, 0) + 1
if speaker_id not in speaker_words:
speaker_words[speaker_id] = []
speaker_words[speaker_id].append(word_obj.text)

if speaker_word_counts:
dominant_speaker = max(speaker_word_counts, key=speaker_word_counts.get)
return " ".join(speaker_words[dominant_speaker])

return transcription.text

elif model_name.startswith("revai/"):
Expand Down