adirik/hierspeechpp 🔢📝🖼️❓✓ → 🖼️

▶️ 4.7K runs 📅 Dec 2023 ⚙️ Cog 0.9.0-beta10 🔗 GitHub 📄 Paper ⚖️ License
audio-to-audio text-to-speech voice-cloning

About

Zero-shot speech synthesizer for text-to-speech and voice conversion

Example Output

Output

Example output

Performance Metrics

3.19s Prediction Time
95.60s Total Time
All Input Parameters
{
  "input_text": "And lay me down in my cold bed and leave my shining lot.",
  "target_voice": "https://replicate.delivery/pbxt/K30ke0FQUcGCa4gQdyhEPaEeGvEwDZmEK3SMtXaoujJSMlSE/reference_1.wav",
  "denoise_ratio": 0,
  "output_sample_rate": 16000,
  "scale_output_volume": false,
  "text_to_vector_temperature": 0.33,
  "voice_conversion_temperature": 0.33
}
Input Parameters
seed Type: integer
Random seed to use for reproducibility.
input_text Type: string
Text input to the model. If provided, it will be used for the speech content of the output.
input_sound Type: string
Sound input to the model in .wav format. If provided, it will be used for the speech content of the output.
target_voice (required) Type: string
A voice clip in .wav format containing the speaker to synthesize.
denoise_ratio Type: numberDefault: 0Range: 0 - 1
Noise control. 0 means no noise reduction, 1 means maximum noise reduction. If noise reduction is desired, it is recommended to set this value to 0.6~0.8
output_sample_rate Default: 16000
Sample rate of the output audio file.
scale_output_volume Type: booleanDefault: false
Scale normalization. If set to true, the output audio will be scaled according to the input sound if provided.
text_to_vector_temperature Type: numberDefault: 0.33Range: 0 - 1
Temperature for text-to-vector model. Larger value corresponds to slightly more random output.
voice_conversion_temperature Type: numberDefault: 0.33Range: 0 - 1
Temperature for the voice conversion model. Larger value corresponds to slightly more random output.
Output Schema

Output

Type: stringFormat: uri

Version Details
Version ID
ff5bcc71dc2c44662291fc348b9ca2eb40107c9f4b377b169fc0dea950c388c8
Version Created
December 14, 2023
Run on Replicate →