rustyverse/rustytorch
2
0
Fork 0
Code Issues Pull Requests Actions 5 Packages Projects Releases Wiki Activity
403 Commits 1 Branch 0 Tags
c7d28705803351e500390c841a95a7f98c1de031
Commit Graph
2 Commits
Author SHA1 Message Date
osobhandClaude Opus 4.7 a5cedfb46a rtx-csm: emotional_speech_guide — CREMA-D vs RAVDESS firdhokk verdict 
8-gen bench (4 emotions × 2 corpora) at seed=42 against firdhokk
Whisper-LV3:

  target    RAVDESS              CREMA-D
  happy     happy (0.999) ✓      happy (0.999) ✓
  angry     neutral (0.92)       sad (0.99)
  fearful   happy (0.998)        fearful (0.984) ✓
  sad       angry (0.99)         fearful (0.99)

CREMA-D 2/4 vs RAVDESS 1/4. Larger / more naturalistic corpus
produces more class-pure fearful direction. Neither corpus solves
angry or sad — recipe shifts into 'vague expressivity' rather than
class-specific corners.

Practical: prefer CREMA-D when available; A/B both per emotion if
class precision matters.

Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>
 2026-04-30 00:01:02 -07:00
osobhandClaude Opus 4.7 5bdb1007af rtx-csm: tts_server_bench — end-to-end latency + concurrency stats 
Sequential and concurrent benchmark client for the tts_server endpoints.
Reports per-endpoint p50/p95/min/max and a serialization-factor metric
for the concurrent /v1/tts case.

First numbers on Metal (M-series, max_audio_ms=2000):
  /health           1.9 ms
  /v1/tts           p50=5.6s p95=6.9s mean=6.0s  (~3x realtime, watermarked)
  /v1/detect        p50=43ms p95=48ms             (decoded=0xCAFE, 16/16 bits)
  /v1/speaker_embed p50=59ms p95=87ms             (100M-param WavLM-SV)
  /v1/speaker_compare p50=100ms p95=108ms         (self-compare cosine=1.0)
  concurrent /v1/tts (n=2): wall=10.2s, serial=14.4s
    serialization factor = 1.42x (Mutex-bound on inference, post-process
    runs in parallel)

Adds reqwest as a dev-dependency (rustls-tls, multipart, json features).

Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>
 2026-04-26 03:33:17 -07:00