Add CXX API for supertonic TTS

[Wasser1462]

Summary by CodeRabbit

• New Features
  • Extended text-to-speech capabilities with support for the Supertonic TTS model, providing an additional model option for speech synthesis.
  • Added a new example demonstrating English speech synthesis using Supertonic with configurable generation parameters and progress tracking.

[coderabbitai]

Caution

Review failed

Pull request was closed or merged during review

📝 Walkthrough

Walkthrough

This PR extends the sherpa-onnx C/C++ API to support the Supertonic TTS model by introducing new configuration structures, populating these fields through the API layers, adding a CMake build target, and providing a complete C++ example demonstrating Supertonic English TTS synthesis with progress callbacks.

Changes

Cohort / File(s)	Summary
Supertonic Model Configuration (Headers) sherpa-onnx/c-api/c-api.h, sherpa-onnx/c-api/cxx-api.h	Added new SherpaOnnxOfflineTtsSupertonicModelConfig and OfflineTtsSupertonicModelConfig structs with seven model component paths (duration_predictor, text_encoder, vector_estimator, vocoder, tts_json, unicode_indexer, voice_style), extended respective OfflineTtsModelConfig structs with supertonic field.
API Implementation sherpa-onnx/c-api/c-api.cc, sherpa-onnx/c-api/cxx-api.cc	Populated new Supertonic model fields from input configuration in GetOfflineTtsConfig and OfflineTts::Create; enhanced JSON value handling in SherpaOnnxOfflineTtsGenerateInternal to conditionally convert non-string values using dump() vs. direct string assignment.
Build Configuration cxx-api-examples/CMakeLists.txt	Added supertonic-tts-en-cxx-api executable target with dependency linking to sherpa-onnx-cxx-api, mirroring existing TTS example structure.
Supertonic TTS Example cxx-api-examples/supertonic-tts-en-cxx-api.cc	New C++ example demonstrating Supertonic English TTS usage: configures OfflineTts with model paths and parameters, defines ProgressCallback for generation progress reporting, synthesizes text to audio via GenerationConfig, and persists output to WAV file.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Possibly related issues

• [Help needed] Refactor the TTS Generate interface #3084: Introduces GenerationConfig usage in TTS generation, which is leveraged by the new Supertonic example code.

Possibly related PRs

• Add C API for Omnilingual ASR CTC models #2773: Follows similar pattern of adding new per-model configuration structs and populating them across C/C++ API layers (ASR variant).
• Add CXX API for KittenTTS #2469: Adds TTS model support (Kitten) by extending OfflineTtsModelConfig and initializing model-specific fields in OfflineTts::Create.
• Add Python API examples for Supertonic TTS #3264: Directly adds Supertonic TTS support across multiple language API bindings.

Suggested labels

size:M

Suggested reviewers

• csukuangfj

Poem

🐰 A new voice joins the chorus bright,
Supertonic sings with ONNX might,
Through C and C++, the configs flow,
Duration, text, and vocoders glow—
Another TTS example hops in delight!

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarizes the main objective of the pull request, which adds C++ API support for Supertonic TTS across multiple files and configuration structures.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

Tip

Try Coding Plans. Let us write the prompt for your AI agent so you can ship faster (with fewer bugs).
Share your feedback on Discord.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request integrates the Supertonic text-to-speech model into the existing framework, enhancing the system's speech synthesis capabilities. It introduces the necessary configuration structures and API mappings to support the new model, alongside a practical C++ example to demonstrate its usage and features like progress callbacks and generation parameter adjustments.

Highlights

• New TTS Model Support: Extended text-to-speech capabilities by adding support for the Supertonic TTS model, providing an additional option for speech synthesis.
• New C++ Example: Added a new C++ example demonstrating English speech synthesis using the Supertonic model, including configurable generation parameters and progress tracking.

Changelog

• cxx-api-examples/CMakeLists.txt
  • Added a new executable target for the Supertonic TTS English CXX API example.
• cxx-api-examples/supertonic-tts-en-cxx-api.cc
  • Introduced a new C++ example for English text-to-speech using the Supertonic model, including configurable generation parameters and progress tracking.
• sherpa-onnx/c-api/c-api.cc
  • Configured the Supertonic model parameters within the C API's OfflineTtsConfig initialization.
  • Improved the JSON parsing logic for extra generation configuration to correctly handle string values, preventing incorrect serialization.
• sherpa-onnx/c-api/c-api.h
  • Defined a new C API struct SherpaOnnxOfflineTtsSupertonicModelConfig to encapsulate Supertonic TTS model parameters.
  • Integrated the SherpaOnnxOfflineTtsSupertonicModelConfig into the SherpaOnnxOfflineTtsModelConfig struct.
• sherpa-onnx/c-api/cxx-api.cc
  • Mapped Supertonic model configuration parameters from the C++ API's OfflineTtsConfig to the underlying C API structure during OfflineTts::Create.
• sherpa-onnx/c-api/cxx-api.h
  • Defined a new C++ API struct OfflineTtsSupertonicModelConfig to hold Supertonic TTS model parameters.
  • Integrated the OfflineTtsSupertonicModelConfig into the OfflineTtsModelConfig C++ struct.

Activity

• The pull request description includes an auto-generated summary by CodeRabbit, detailing new features related to Supertonic TTS support and an example.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request adds CXX API support for the Supertonic TTS model. The changes include adding new configuration structures for Supertonic in both the C and CXX APIs, implementing the logic to handle these configurations, and providing a new example file to demonstrate its usage. The implementation looks solid and consistent with the existing structure for other TTS models. I've included one suggestion to improve the maintainability of the new example file by refactoring hardcoded model paths.

[gemini-code-assist]

To improve maintainability and reduce redundancy, consider defining the base path for the model files in a constant string and then constructing the full paths from it. This makes it easier to update the model path in the future.

Suggested change

	config.model.supertonic.duration_predictor =
	"./sherpa-onnx-supertonic-tts-int8-2026-03-06/"
	"duration_predictor.int8.onnx";
	config.model.supertonic.text_encoder =
	"./sherpa-onnx-supertonic-tts-int8-2026-03-06/text_encoder.int8.onnx";
	config.model.supertonic.vector_estimator =
	"./sherpa-onnx-supertonic-tts-int8-2026-03-06/vector_estimator.int8.onnx";
	config.model.supertonic.vocoder =
	"./sherpa-onnx-supertonic-tts-int8-2026-03-06/vocoder.int8.onnx";
	config.model.supertonic.tts_json =
	"./sherpa-onnx-supertonic-tts-int8-2026-03-06/tts.json";
	config.model.supertonic.unicode_indexer =
	"./sherpa-onnx-supertonic-tts-int8-2026-03-06/unicode_indexer.bin";
	config.model.supertonic.voice_style =
	"./sherpa-onnx-supertonic-tts-int8-2026-03-06/voice.bin";
	const std::string model_dir = "./sherpa-onnx-supertonic-tts-int8-2026-03-06/";
	config.model.supertonic.duration_predictor =
	model_dir + "duration_predictor.int8.onnx";
	config.model.supertonic.text_encoder = model_dir + "text_encoder.int8.onnx";
	config.model.supertonic.vector_estimator =
	model_dir + "vector_estimator.int8.onnx";
	config.model.supertonic.vocoder = model_dir + "vocoder.int8.onnx";
	config.model.supertonic.tts_json = model_dir + "tts.json";
	config.model.supertonic.unicode_indexer =
	model_dir + "unicode_indexer.bin";
	config.model.supertonic.voice_style = model_dir + "voice.bin";