From 9f0b18b03d8e442c90c29ec27ef987b0878de8db Mon Sep 17 00:00:00 2001 From: sathwika Date: Tue, 7 Apr 2026 18:18:55 +0530 Subject: [PATCH] Add changelog fragments for PR #4249 --- changelog/4249.added.md | 9 +++++++++ changelog/4249.changed.md | 5 +++++ 2 files changed, 14 insertions(+) create mode 100644 changelog/4249.added.md create mode 100644 changelog/4249.changed.md diff --git a/changelog/4249.added.md b/changelog/4249.added.md new file mode 100644 index 000000000..516e2e3a8 --- /dev/null +++ b/changelog/4249.added.md @@ -0,0 +1,9 @@ +- Added cross-sentence stitching support to NvidiaTTSService. Multiple sentences within an LLM turn are fed into a single `SynthesizeOnline` gRPC stream, enabling Magpie's seamless audio across sentence boundaries (requires Magpie TTS model v1.7.0+). +- Added new parameters to NvidiaTTSService: + - `custom_dictionary` for IPA-based custom pronunciation + - `encoding` for output audio encoding format + - `zero_shot_audio_prompt_file` for zero-shot voice cloning with file caching + - `audio_prompt_encoding` for zero-shot audio prompt format +- Added `can_generate_metrics` returning True for NvidiaTTSService. +- Added `stop_all_metrics()` call on audio context interruption. +- Added gRPC error handling for synthesis config retrieval. \ No newline at end of file diff --git a/changelog/4249.changed.md b/changelog/4249.changed.md new file mode 100644 index 000000000..460a06afa --- /dev/null +++ b/changelog/4249.changed.md @@ -0,0 +1,5 @@ +- Updated NvidiaTTSService: + - Made `api_key` optional for local NIM deployments + - Fixed `_update_settings` to correctly indicate that runtime settings updates (voice, language, quality) take effect on the next turn + - Replaced per-sentence synchronous `synthesize_online` calls with async queue-backed gRPC streaming + - Renamed Riva references to Nemotron Speech \ No newline at end of file