tzhouamandClaude Fable 5.1
49a839ecf9
[Edge][WP5] Add codec-token streaming mode with on-device decoding
* qwen3_tts_talker_only pipeline (single terminal LLM_AR stage,
final_output_type "latent") + vllm_omni/deploy/qwen3_tts_talker_only.yaml;
a terminal talker publishes its per-step codes.audio rows to the client
(omni_client_multimodal_output_keys).
* /v1/audio/speech/stream: session.config.output_mode="codec_tokens" streams
codec.start (JSON) + binary frames (<IHB seq/n_frames/flags + int16 rows,
vllm_omni/entrypoints/openai/codec_stream.py) + codec.done; new input.cancel
message aborts an in-flight generation for both codec and PCM sessions
(single reader task + session inbox replaces direct socket reads); optional
session.config.seed; audio.done gains "cancelled" only when it happened.
* serving_speech._generate_codec_chunks aligns the cumulative code tensor to
output token ids (drops prefill placeholder/EOS rows) and emits deltas.
* TTS adapters: supports_codec_stream / codec_stream_spec (Qwen3-TTS: 16
codebooks, 2048 ids, 12.5 Hz, 24 kHz, decoder qwen3_tts_tokenizer_v2_12hz).
* examples/online_serving/text_to_speech/qwen3_tts/edge_decoder_client.py
decodes locally with Qwen3TTSTokenizerV2Decoder (ramp schedule, 72-frame
context, cancel/reset), codec_parity_check.py and
benchmarks/tts/codec_stream_e2e.py compare against the two-stage server.
* docs/serving/codec_token_streaming.md; CPU unit tests for the wire format,
alignment, session config and handler (frames, unsupported adapter, cancel);
L3 e2e tests/e2e/online_serving/test_qwen3_tts_codec_stream.py.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-09 11:10:13 +08:00
..
2026-07-27 23:35:49 +08:00
2026-09-05 08:20:21 +00:00
2026-09-05 08:20:21 +00:00
2026-09-09 11:10:13 +08:00
2026-09-06 22:25:35 +08:00
2026-09-09 11:10:13 +08:00
2026-09-07 14:27:24 +00:00
2026-09-06 22:25:35 +08:00
2026-09-09 11:01:40 +08:00
2026-09-09 11:01:39 +08:00
2026-09-09 11:10:13 +08:00
2026-09-07 13:38:23 +08:00
2026-08-28 09:30:17 +08:00
2026-09-07 11:40:04 +08:00
2026-01-27 10:22:40 +08:00
2026-08-29 08:59:05 -07:00
2026-09-09 11:10:13 +08:00
2026-09-05 15:32:30 +08:00
2026-09-05 08:20:21 +00:00
2026-09-09 11:02:14 +08:00
2026-07-11 18:48:43 +00:00
2026-04-24 16:30:24 +08:00
2026-09-06 04:48:38 +00:00
2026-06-10 21:03:19 +00:00
2025-09-11 15:43:43 +08:00
2026-08-28 08:32:32 +08:00
2026-09-04 14:44:10 -07:00
2026-08-22 14:26:49 +08:00
2026-09-09 11:01:39 +08:00
2026-07-27 23:35:49 +08:00
2026-09-03 18:49:19 -07:00
2026-06-11 21:09:24 +00:00
2025-12-15 07:13:36 +00:00
2026-09-09 11:02:14 +08:00
2026-08-27 06:12:55 +00:00
2026-04-17 20:15:16 +08:00