rixty
/
VoiceCraft
mirror of https://github.com/jasonppy/VoiceCraft.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

add short form commands

This commit is contained in:
Pranay Gosar 2024-04-23 18:55:34 -05:00
parent 59877c085e
commit 1850da9210
2 changed files with 53 additions and 50 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

58
inference_demo.py
View File

@ -26,44 +26,45 @@ def parse_arguments():
parser = argparse.ArgumentParser( parser = argparse.ArgumentParser(
description="VoiceCraft TTS Inference: see the script for more information on the options") description="VoiceCraft TTS Inference: see the script for more information on the options")
parser.add_argument("--model_name", type=str, default="giga330M.pth", choices=[ parser.add_argument("-m", "--model_name", type=str, default="giga330M.pth", choices=[
"giga330M.pth", "gigaHalfLibri330M_TTSEnhanced_max16s.pth", "giga830M.pth"], "giga330M.pth", "gigaHalfLibri330M_TTSEnhanced_max16s.pth", "giga830M.pth"],
help="VoiceCraft model to use") help="VoiceCraft model to use")
parser.add_argument("--codec_audio_sr", type=int, parser.add_argument("-st", "--silence_tokens", type=int, nargs="*",
default=16000, help="Audio sampling rate for the codec")
parser.add_argument("--codec_sr", type=int, default=50,
help="Sampling rate for the codec")
parser.add_argument("--top_k", type=float, default=0,
help="Top-k value")
parser.add_argument("--top_p", type=float, default=0.9,
help="Top-p value")
parser.add_argument("--temperature", type=float,
default=1.0, help="Temperature for sampling")
parser.add_argument("--silence_tokens", type=int, nargs="*",
default=[1388, 1898, 131], help="Silence token IDs") default=[1388, 1898, 131], help="Silence token IDs")
parser.add_argument("--kvcache", type=int, default=1, choices=[0, 1], parser.add_argument("-casr", "--codec_audio_sr", type=int,
help="Key-value cache flag (0 or 1)") default=16000, help="Codec audio sample rate.")
parser.add_argument("--stop_repetition", type=int, parser.add_argument("-csr", "--codec_sr", type=int, default=50,
default=3, help="Stop repetition for generation") help="Codec sample rate.")
parser.add_argument("-k", "--top_k", type=float,
default=0, help="Top k value.")
parser.add_argument("-p", "--top_p", type=float,
default=0.8, help="Top p value.")
parser.add_argument("-t", "--temperature", type=float,
default=1, help="Temperature value.")
parser.add_argument("-kv", "--kvcache", type=float, choices=[0, 1],
default=0, help="Kvcache value.")
parser.add_argument("-sr", "--stop_repetition", type=int,
default=-1, help="Stop repetition for generation")
parser.add_argument("--sample_batch_size", type=int, parser.add_argument("--sample_batch_size", type=int,
default=3, help="Batch size for sampling") default=3, help="Batch size for sampling")
parser.add_argument("--seed", type=int, default=1, parser.add_argument("-s", "--seed", type=int,
help="Random seed for reproducibility") default=1, help="Seed value.")
parser.add_argument("--beam_size", type=int, default=10, parser.add_argument("-bs", "--beam_size", type=int, default=10,
help="beam size for MFA alignment") help="beam size for MFA alignment")
parser.add_argument("--retry_beam_size", type=int, default=40, parser.add_argument("-rbs", "--retry_beam_size", type=int, default=40,
help="retry beam size for MFA alignment") help="retry beam size for MFA alignment")
parser.add_argument("--output_dir", type=str, default="./generated_tts", parser.add_argument("--output_dir", type=str, default="./generated_tts",
help="directory to save generated audio") help="directory to save generated audio")
parser.add_argument("--original_audio", type=str, parser.add_argument("-oa", "--original_audio", type=str,
default="./demo/84_121550_000074_000000.wav", help="location of target audio file") default="./demo/84_121550_000074_000000.wav", help="location of audio file")
parser.add_argument("--original_transcript", type=str, parser.add_argument("-ot", "--original_transcript", type=str,
default="But when I had approached so near to them The common object, which the sense deceives, Lost not by distance any of its marks,", default="But when I had approached so near to them The common object, which the sense deceives, Lost not by distance any of its marks,",
help="original audio transcript") help="original transcript")
parser.add_argument("--target_transcript", type=str, parser.add_argument("-tt", "--target_transcript", type=str,
default="Gwynplaine had, besides, for his work and for his feats of strength, I cannot believe that the same model can also do text to speech synthesis too!", default="But when I saw the mirage of the lake in the distance, which the sense deceives, Lost not by distance any of its marks,",
help="target audio transcript") help="target transcript")
parser.add_argument("--cut_off_sec", type=float, default=3.6, parser.add_argument("-co", "--cut_off_sec", type=float, default=3.6,
help="cut off point in seconds for input prompt") help="cut off point in seconds for input prompt")
args = parser.parse_args() args = parser.parse_args()
return args return args
@ -145,6 +146,7 @@ audio_dur = info.num_frames / info.sample_rate
assert cut_off_sec < audio_dur, f"cut_off_sec {cut_off_sec} is larger than the audio duration {audio_dur}" assert cut_off_sec < audio_dur, f"cut_off_sec {cut_off_sec} is larger than the audio duration {audio_dur}"
prompt_end_frame = int(cut_off_sec * info.sample_rate) prompt_end_frame = int(cut_off_sec * info.sample_rate)
def seed_everything(seed): def seed_everything(seed):
os.environ['PYTHONHASHSEED'] = str(seed) os.environ['PYTHONHASHSEED'] = str(seed)
random.seed(seed) random.seed(seed)

45
speech_editing_demo.py
View File

@ -27,46 +27,47 @@ def parse_arguments():
parser = argparse.ArgumentParser( parser = argparse.ArgumentParser(
description="VoiceCraft Speech Editing: see the script for more information on the options") description="VoiceCraft Speech Editing: see the script for more information on the options")
parser.add_argument("--model_name", type=str, default="giga330M.pth", choices=[ parser.add_argument("-m", "--model_name", type=str, default="giga330M.pth", choices=[
"giga330M.pth", "gigaHalfLibri330M_TTSEnhanced_max16s.pth", "giga830M.pth"], "giga330M.pth", "gigaHalfLibri330M_TTSEnhanced_max16s.pth", "giga830M.pth"],
help="VoiceCraft model to use") help="VoiceCraft model to use")
parser.add_argument("--silence_tokens", type=int, nargs="*", parser.add_argument("-st", "--silence_tokens", type=int, nargs="*",
default=[1388, 1898, 131], help="Silence token IDs") default=[1388, 1898, 131], help="Silence token IDs")
parser.add_argument("--left_margin", type=float, parser.add_argument("-lm", "--left_margin", type=float,
default=0.08, help="Left margin value.") default=0.08, help="Left margin value.")
parser.add_argument("--right_margin", type=float, parser.add_argument("-rm", "--right_margin", type=float,
default=0.08, help="Right margin value.") default=0.08, help="Right margin value.")
parser.add_argument("--codec_audio_sr", type=int, parser.add_argument("-casr", "--codec_audio_sr", type=int,
default=16000, help="Codec audio sample rate.") default=16000, help="Codec audio sample rate.")
parser.add_argument("--codec_sr", type=int, default=50, parser.add_argument("-csr", "--codec_sr", type=int, default=50,
help="Codec sample rate.") help="Codec sample rate.")
parser.add_argument("--top_k", type=float, default=0, help="Top k value.") parser.add_argument("-k", "--top_k", type=float,
parser.add_argument("--top_p", type=float, default=0, help="Top k value.")
parser.add_argument("-p", "--top_p", type=float,
default=0.8, help="Top p value.") default=0.8, help="Top p value.")
parser.add_argument("--temperature", type=float, parser.add_argument("-t", "--temperature", type=float,
default=1, help="Temperature value.") default=1, help="Temperature value.")
parser.add_argument("--kvcache", type=float, parser.add_argument("-kv", "--kvcache", type=float, choices=[0, 1],
default=0, help="Kvcache value.") default=0, help="Kvcache value.")
parser.add_argument("--seed", type=int, default=1, help="Seed value.") parser.add_argument("-sr", "--stop_repetition", type=int,
parser.add_argument("--beam_size", type=int, default=10,
help="beam size for MFA alignment")
parser.add_argument("--retry_beam_size", type=int, default=40,
help="retry beam size for MFA alignment")
parser.add_argument("--original_audio", type=str,
default="./demo/84_121550_000074_000000.wav", help="location of audio file")
parser.add_argument("--stop_repetition", type=int,
default=-1, help="Stop repetition for generation") default=-1, help="Stop repetition for generation")
parser.add_argument("--original_transcript", type=str, parser.add_argument("-s", "--seed", type=int, default=1, help="Seed value.")
parser.add_argument("-bs", "--beam_size", type=int, default=10,
help="beam size for MFA alignment")
parser.add_argument("-rbs", "--retry_beam_size", type=int, default=40,
help="retry beam size for MFA alignment")
parser.add_argument("-oa", "--original_audio", type=str,
default="./demo/84_121550_000074_000000.wav", help="location of audio file")
parser.add_argument("-ot", "--original_transcript", type=str,
default="But when I had approached so near to them The common object, which the sense deceives, Lost not by distance any of its marks,", default="But when I had approached so near to them The common object, which the sense deceives, Lost not by distance any of its marks,",
help="original transcript") help="original transcript")
parser.add_argument("--target_transcript", type=str, parser.add_argument("-tt", "--target_transcript", type=str,
default="But when I saw the mirage of the lake in the distance, which the sense deceives, Lost not by distance any of its marks,", default="But when I saw the mirage of the lake in the distance, which the sense deceives, Lost not by distance any of its marks,",
help="target transcript") help="target transcript")
parser.add_argument("--edit_type", type=str, parser.add_argument("-et", "--edit_type", type=str,
default="substitution", default="substitution",
choices=["insertion", "substitution", "deletion"], choices=["insertion", "substitution", "deletion"],
help="type of specified edit") help="type of specified edit")
parser.add_argument("--output_dir", type=str, parser.add_argument("-o", "--output_dir", type=str,
default="./demo/generated_se", help="output directory") default="./demo/generated_se", help="output directory")
args = parser.parse_args() args = parser.parse_args()
return args return args