# Soul-AILab/SoulX-Podcast

SoulX-Podcast is an inference codebase by the Soul AI team for generating high-fidelity podcasts from text.

Repository: https://github.com/Soul-AILab/SoulX-Podcast
Canonical: https://ross.abutalabs.com/products/soulx-podcast
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2025-12-11T12:48:46+00:00

## Health v2 (maintenance only)
Score: 42/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 56, release rhythm 35, longevity 24
- inputs: {"age_days": 345, "days_push": 265, "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 3535, forks 457 (observed 2026-08-28T04:08:09.043223+00:00)

## What it is
SoulX-Podcast is the official inference codebase for a text-to-speech model that generates long-form, multi-turn, multi-speaker podcast-style dialogic speech. It supports Mandarin, English, and several Chinese dialects with zero-shot voice cloning and paralinguistic controls like laughter and sighs.

## Use cases
- generate realistic podcast audio from text scripts
- synthesize multi-speaker dialogue speech
- clone voices across Chinese dialects
- add laughter and sighs to synthesized speech
- convert long-form text into natural spoken audio
- build an AI podcast generator

## When to choose
- you need long-form multi-speaker dialogic speech generation
- you need Chinese dialect support like Sichuanese or Cantonese
- you want paralinguistic realism such as laughter and breathing in TTS output
- you need zero-shot voice cloning for podcast-style content

## When to avoid
- you only need simple single-phrase TTS with minimal setup
- you need languages beyond Mandarin, English, and Chinese dialects
- you lack GPU resources for model inference
- you need a production SaaS rather than an inference codebase

## Facets
- artifact type: library
- maturity: active
- function: tts, speech-recognition, machine-learning, llm-inference
- domain: speech-processing, artificial-intelligence
- platform: python
- tags: podcast-generation, voice-cloning, multi-speaker-speech, chinese-dialects, paralinguistic-control, zero-shot-tts, audio, natural-language-processing, gpu, linux

## Member repositories
- Soul-AILab/SoulX-Podcast (main) score 42

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:09.043223+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 2026-08-29T18:35:03.183111+00:00, confidence not recorded.
  - readme: https://github.com/Soul-AILab/SoulX-Podcast (fetched 2026-08-28T04:08:09.043223+00:00, sha d244aace08a3)
- Data as of 2026-08-30T08:39:29.467469+00:00.