# WhisperSpeech/WhisperSpeech

An Open Source text-to-speech system built by inverting Whisper.

Repository: https://github.com/WhisperSpeech/WhisperSpeech
Canonical: https://ross.abutalabs.com/products/whisperspeech
Homepage: https://whisperspeech.github.io/WhisperSpeech/
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: pytorch, speech-synthesis, tts
Last push: 2025-12-14T11:11:47+00:00

## Health v2 (maintenance only)
Score: 56/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 57, release rhythm 35, longevity 92
- inputs: {"age_days": 1296, "days_push": 262, "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 4639, forks 275 (observed 2026-08-28T04:08:55.459800+00:00)

## What it is
WhisperSpeech is an open-source text-to-speech system built by inverting OpenAI's Whisper model, aiming to be 'Stable Diffusion for speech'. It supports voice cloning, multilingual synthesis, and runs faster than real time on consumer GPUs, with all code and training data properly licensed for commercial use.

## Use cases
- generate speech from text with an open-source model
- clone a voice from a short audio sample
- synthesize multilingual or code-switching sentences
- run fast TTS locally on a consumer GPU
- build a commercially safe TTS product without licensing risk
- fine-tune or hack a speech synthesis pipeline

## When to choose
- you need open-source, commercially usable TTS with permissive licenses
- you want voice cloning and multilingual support
- you have a GPU and want faster-than-real-time local synthesis
- you want a hackable research-friendly speech model

## When to avoid
- you need production TTS with many polished prebuilt voices out of the box
- you have no GPU and need low-latency CPU inference
- you need broad language coverage today (English is the main current release)

## Facets
- artifact type: library
- maturity: active
- function: tts, machine-learning, deep-learning, audio-processing
- domain: speech-processing, artificial-intelligence, deep-learning
- platform: python, cross-platform
- tags: speech-synthesis, voice-cloning, pytorch, whisper, open-source-models, audio-generation, audio, gpu

## Member repositories
- WhisperSpeech/WhisperSpeech (main) score 56

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:55.459800+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 2026-08-29T18:19:33.474018+00:00, confidence not recorded.
  - readme: https://github.com/WhisperSpeech/WhisperSpeech (fetched 2026-08-28T04:08:55.459800+00:00, sha 817d9e061f9f)
  - homepage: https://whisperspeech.github.io/WhisperSpeech/ (fetched 2026-08-29T09:04:14.379571+00:00, sha 3b7140763a04)
- Data as of 2026-08-30T08:39:29.467469+00:00.