jaywalnut310/glow-tts

• highreadme#1

  Reposition README opening to clarify project type and audience

  Why:

  CURRENT

  In our recent paper, we propose Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Search. Recently, text-to-speech (TTS) models such as FastSpeech and ParaNet have been proposed to generate mel-spectrograms from text in parallel. Despite the advantage, the parallel TTS models cannot be trained without guidance from autoregressive TTS models as their external aligners. In this work, we propose Glow-TTS, a flow-based generative model for parallel TTS that does not require any external aligner. By combining the properties of flows and dynamic programming, the proposed model searches for the most probable monotonic alignment between text and the latent representation of speech on its own. We demonstrate that enforcing hard monotonic alignments enables robust TTS, which generalizes to long utterances, and employing generative flows enables fast, diverse, and controllable speech synthesis. Glow-TTS obtains an order-of-magnitude speed-up over the autoregressive model, Tacotron 2, at synthesis with comparable speech quality. We further show that our model can be easily extended to a multi-speaker setting.

  COPY-PASTE FIX

  Copy

  Glow-TTS is an open-source research implementation of a **generative flow-based model for parallel Text-to-Speech (TTS)**, designed for researchers and developers. It enables fast, diverse, and controllable speech synthesis by learning monotonic alignments without requiring external aligners, offering an order-of-magnitude speed-up over autoregressive models like Tacotron 2.

  OpenDoneDismiss
• highcomparison#2

  Add a "Comparison to Alternatives" section in the README

  Why:

  COPY-PASTE FIX

  Copy

  ## Comparison to Alternatives
  
  Glow-TTS is an open-source research project, offering a distinct approach compared to commercial Text-to-Speech APIs (e.g., Google Cloud TTS, Amazon Polly, ElevenLabs) by providing full control and transparency for developers and researchers. Unlike many parallel TTS models that rely on external aligners, Glow-TTS leverages a flow-based generative model with monotonic alignment search, ensuring robust and fast speech synthesis without such dependencies.

  OpenDoneDismiss
• mediumtopics#3

  Add more specific keywords to repository topics

  Why:

  CURRENT

  deep-learning, pytorch, speech-synthesis, text-to-speech, tts

  COPY-PASTE FIX

  Copy

  deep-learning, pytorch, speech-synthesis, text-to-speech, tts, parallel-tts, non-autoregressive, generative-model, flow-based

  OpenDoneDismiss

Category GEO backends resolved for this scan: google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• CATEGORY QUERY

  How to generate high-quality speech from text quickly without an external aligner?

  you: not recommended
  AI recommended (in order):

  1. Google Cloud Text-to-Speech
  2. Amazon Polly
  3. Microsoft Azure Text to Speech
  4. ElevenLabs
  5. OpenAI TTS
  6. Meta Voicebox

  AI recommended 6 alternatives but never named jaywalnut310/glow-tts. This is the gap to close.

  Show full AI answer
• CATEGORY QUERY

  Looking for a parallel text-to-speech model that offers diverse and controllable output.

  you: #11
  AI recommended (in order):

  1. VALL-E X
  2. Meta Voice
  3. Voicebox
  4. YourTTS
  5. StyleTTS 2
  6. FastSpeech 2
  7. FastSpeech 2s
  8. Tacotron 2
  9. WaveGlow
  10. Hifi-GAN
  11. Glow-TTS ← you
  Show full AI answer

• Metadata completeness

  warn

  Suggestion:

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of jaywalnut310/glow-tts?

  pass
  ✓AI named jaywalnut310/glow-tts explicitly

  AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

• If a team adopts jaywalnut310/glow-tts in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI named jaywalnut310/glow-tts explicitly

  AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

• In one sentence, what problem does the repo jaywalnut310/glow-tts solve, and who is the primary audience?

  pass
  ⚠AI did not name jaywalnut310/glow-tts — likely talking about a different project

  AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?