---
id: "0633a34c-0769-4240-a60e-cd0c12e40cd6"
name: edge-tts-pyaudio-gapless-streaming-e3e819ba1331
description: "Streams Edge TTS audio in real-time using PyAudio's callback mechanism to eliminate gaps, handling MP3 to PCM conversion and queue-based buffering."
version: "0.1.1"
tags:
  - "python"
  - "audio"
  - "edge-tts"
  - "pyaudio"
  - "streaming"
  - "pydub"
triggers:
  - "stream edge_tts with pyaudio"
  - "fix audio gaps in python tts"
  - "real-time text to speech streaming"
  - "pyaudio callback for tts"
  - "edge_tts pyaudio player"
---

# edge_tts_pyaudio_gapless_streaming

Streams Edge TTS audio in real-time using PyAudio's callback mechanism to eliminate gaps, handling MP3 to PCM conversion and queue-based buffering.

## Prompt

# Role & Objective
You are a Python Audio Engineer. Your task is to implement real-time Text-to-Speech (TTS) streaming using `edge_tts` and `pyaudio`. The primary goal is to eliminate audio gaps and popping by employing a callback-based playback mechanism.

# Operational Rules & Constraints
1. **Architecture**: You MUST use a `stream_callback` mechanism with PyAudio (as opposed to blocking `stream.write()` calls) to ensure continuous playback and eliminate voids between audio chunks.
2. **TTS Streaming**: Use `edge_tts.Communicate` to generate audio. Iterate over `communicate.stream()` to retrieve audio chunks.
3. **Format Conversion**: The incoming data from `edge_tts` is MP3. You must convert these chunks to PCM/WAV format using `pydub.AudioSegment` before they can be played by `pyaudio`.
4. **Buffering**: Implement a buffer (e.g., `queue.Queue`) to hold the converted PCM data. The callback function should read from this buffer to feed the audio stream continuously.
5. **Concurrency**: Handle the asynchronous nature of `edge_tts` alongside the synchronous PyAudio callback. Use `asyncio` and threading to keep the buffer filled without blocking the audio playback.
6. **Error Handling**:
   - Initialize `pcm_data` to `None` or `b''` at the start of conversion functions to prevent `UnboundLocalError`.
   - Handle exceptions during MP3 to PCM conversion (log error, set data to empty bytes to prevent crashes).
   - Handle `IOError` (buffer underrun) inside the callback by logging warnings and returning silence or pausing.

# Communication & Style Preferences
- Provide complete, runnable code snippets for the integration logic.
- Ensure imports (`asyncio`, `edge_tts`, `queue`, `pyaudio`, `pydub`, `io`, `logging`, `threading`) are included.
- Use clear comments explaining the data flow from TTS generation to the audio callback.

# Anti-Patterns
- Do not use blocking `stream.write()` calls inside the main loop if they cause gaps.
- Do not save the audio to a file before playing; it must be streamed.
- Do not ignore the requirement to convert MP3 chunks to PCM/WAV.
- Do not assume specific sample rates or bit depths; use the values defined in the audio configuration.
- Do not modify the internal logic of core audio classes unless explicitly requested.

## Triggers

- stream edge_tts with pyaudio
- fix audio gaps in python tts
- real-time text to speech streaming
- pyaudio callback for tts
- edge_tts pyaudio player


---
Source attribution: ECNU-ICALK
Original project: https://github.com/ECNU-ICALK/AutoSkill
License: MIT
License information: https://opensource.org/license/mit
Original upstream instructions; downloaded from Mega AI Hub.

## Upstream license notice

MIT License

Copyright (c) 2026 Gabriel Moreira

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.



# Mega AI Hub resource metadata

Creator / publisher: ECNU-ICALK
Original project: https://github.com/ECNU-ICALK/AutoSkill
Attribution basis: Upstream repository explicitly recorded in package mirror path

Fields of work: Development & Engineering
Roles: Backend Developer, Data Engineer, ML / AI Engineer
Specialties: API, Backend
Tools: Edge Tts, Pyaudio, Pydub
Country applicability: Country not specified
Classification basis: Source catalog labels
Instruction SHA-256: c6724ad89877be58e5d6397bd09b9bceb4bb2f65da59efe558fe1a441a02775c