Files

190 lines
6.3 KiB
Python

#!/usr/bin/env python3
"""
ElevenLabs Instant Voice Cloning (IVC) API wrapper.
Usage (CLI):
python3 voiceclone.py --name "MyVoice" --files audio1.mp3 audio2.mp3 --language de
python3 voiceclone.py --name "MyVoice" --files *.m4a --description "German male voice"
Usage (Module):
from voiceclone import clone_voice
voice_id = clone_voice("MyVoice", ["audio1.mp3", "audio2.mp3"], language="de")
"""
import argparse
import os
import sys
import requests
from pathlib import Path
def clone_voice(
name: str,
files: list[str],
api_key: str | None = None,
description: str | None = None,
labels: dict[str, str] | None = None,
remove_background_noise: bool = False,
) -> dict:
"""
Create an instant voice clone from audio samples.
Args:
name: Display name for the voice
files: List of paths to audio files (mp3, m4a, wav, etc.)
api_key: ElevenLabs API key (defaults to ELEVENLABS_API_KEY env)
description: Optional description of the voice
labels: Optional dict with keys like 'language', 'accent', 'gender', 'age'
remove_background_noise: Whether to apply audio isolation
Returns:
dict with 'voice_id' and 'requires_verification'
"""
api_key = api_key or os.environ.get("ELEVENLABS_API_KEY")
if not api_key:
raise ValueError("ELEVENLABS_API_KEY not set")
url = "https://api.elevenlabs.io/v1/voices/add"
headers = {
"xi-api-key": api_key,
}
# Prepare multipart form data
form_data = {
"name": (None, name),
"remove_background_noise": (None, str(remove_background_noise).lower()),
}
if description:
form_data["description"] = (None, description)
# Add labels as JSON string if provided
if labels:
import json
form_data["labels"] = (None, json.dumps(labels))
# Add audio files
file_handles = []
try:
for file_path in files:
path = Path(file_path)
if not path.exists():
raise FileNotFoundError(f"Audio file not found: {file_path}")
# Determine MIME type
suffix = path.suffix.lower()
mime_types = {
".mp3": "audio/mpeg",
".m4a": "audio/x-m4a",
".wav": "audio/wav",
".ogg": "audio/ogg",
".flac": "audio/flac",
".webm": "audio/webm",
}
mime_type = mime_types.get(suffix, "audio/mpeg")
fh = open(path, "rb")
file_handles.append(fh)
form_data[f"files"] = (path.name, fh, mime_type)
# Need to use files parameter for multiple files with same key
# Rebuild as list of tuples for requests
files_list = []
for key, value in form_data.items():
if key != "files":
files_list.append((key, value))
# Add all audio files with the same "files" key
for i, file_path in enumerate(files):
path = Path(file_path)
suffix = path.suffix.lower()
mime_types = {
".mp3": "audio/mpeg",
".m4a": "audio/x-m4a",
".wav": "audio/wav",
".ogg": "audio/ogg",
".flac": "audio/flac",
".webm": "audio/webm",
}
mime_type = mime_types.get(suffix, "audio/mpeg")
files_list.append(("files", (path.name, file_handles[i], mime_type)))
response = requests.post(url, headers=headers, files=files_list)
finally:
for fh in file_handles:
fh.close()
if response.status_code != 200:
raise RuntimeError(f"ElevenLabs API error {response.status_code}: {response.text}")
return response.json()
def main():
parser = argparse.ArgumentParser(
description="Create an instant voice clone via ElevenLabs API",
formatter_class=argparse.RawDescriptionHelpFormatter,
epilog="""
Examples:
# Clone from multiple audio files
python3 voiceclone.py --name "Andi" --files audio1.m4a audio2.m4a audio3.m4a
# With language label
python3 voiceclone.py --name "Andi" --files *.m4a --language de --gender male
# With description and noise removal
python3 voiceclone.py --name "Andi" --files *.m4a --description "German male voice" --denoise
""",
)
parser.add_argument("--name", "-n", required=True, help="Name for the cloned voice")
parser.add_argument("--files", "-f", nargs="+", required=True, help="Audio file(s) for cloning")
parser.add_argument("--description", "-d", help="Description of the voice")
parser.add_argument("--language", "-l", help="Language label (e.g., 'de', 'en')")
parser.add_argument("--accent", help="Accent label (e.g., 'german', 'british')")
parser.add_argument("--gender", "-g", help="Gender label ('male' or 'female')")
parser.add_argument("--age", help="Age label (e.g., 'young', 'middle_aged', 'old')")
parser.add_argument("--denoise", action="store_true", help="Remove background noise from samples")
parser.add_argument("--json", action="store_true", help="Output raw JSON response")
args = parser.parse_args()
# Build labels dict from provided options
labels = {}
if args.language:
labels["language"] = args.language
if args.accent:
labels["accent"] = args.accent
if args.gender:
labels["gender"] = args.gender
if args.age:
labels["age"] = args.age
try:
result = clone_voice(
name=args.name,
files=args.files,
description=args.description,
labels=labels if labels else None,
remove_background_noise=args.denoise,
)
if args.json:
import json
print(json.dumps(result, indent=2))
else:
print(f"✅ Voice cloned successfully!")
print(f" Name: {args.name}")
print(f" Voice ID: {result['voice_id']}")
if result.get("requires_verification"):
print(" ⚠️ Voice requires verification")
except Exception as e:
print(f"❌ Error: {e}", file=sys.stderr)
sys.exit(1)
if __name__ == "__main__":
main()