# andabi/deep-voice-conversion

Deep neural networks for voice conversion (voice style transfer) in Tensorflow

Repository: https://github.com/andabi/deep-voice-conversion
Canonical: https://ross.abutalabs.com/products/deep-voice-conversion
Language: Python
License: MIT
License Family: permissive
Last push: 2022-09-30T19:45:52+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3243, "days_push": 1433, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3938, forks 824 (observed 2026-08-28T04:08:30.113981+00:00)

## What it is
A TensorFlow implementation of deep neural networks for voice conversion (voice style transfer) that converts a source speaker's voice into a target speaker's voice without parallel training data. It uses a two-module architecture (phoneme classification plus speech synthesis) with CBHG modules from Tacotron.

## Use cases
- convert my voice to sound like another speaker
- voice style transfer with deep learning
- train a voice conversion model without parallel data
- clone a target speaker's voice from waveforms
- experiment with Tacotron-style speech synthesis
- phoneme classification from spectrograms

## When to choose
- you want to convert speech to a specific target speaker using only target waveforms
- you need a research/educational reference implementation of non-parallel voice conversion
- you work in TensorFlow and want to study CBHG-based speech models

## When to avoid
- you need production-quality, real-time voice conversion
- you want a maintained tool with recent updates or pretrained multi-speaker models
- you need a simple API rather than a research codebase requiring TIMIT and custom datasets

## Facets
- artifact type: library
- maturity: maintenance
- function: deep-learning, speech-recognition, audio-processing, machine-learning
- domain: speech-processing, deep-learning, machine-learning
- platform: python
- tags: voice-conversion, tensorflow, speech-synthesis, style-transfer, non-parallel-data, audio, linux, macos

## Member repositories
- andabi/deep-voice-conversion (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:30.113981+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:24:38.307539+00:00, confidence not recorded.
  - readme: https://github.com/andabi/deep-voice-conversion (fetched 2026-08-28T04:08:30.113981+00:00, sha 87282580c41b)
- Data as of 2026-08-30T08:39:29.467469+00:00.