rustyverse/rustytorch
2
0
Fork 0
Code Issues Pull Requests Actions 14 Packages Projects Releases Wiki Activity

rtx-csm: emotional_speech_guide — CREMA-D vs RAVDESS firdhokk verdict

Browse Source 
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]>
This commit is contained in:
osobh
2026-04-30 00:01:02 -07:00
co-authored by Claude Opus 4.7
parent f4d8268381
commit a5cedfb46a
69 changed files with 1026 additions and 734 deletions
Download Patch File Download Diff File Expand all files Collapse all files
crates/models/rtx-csm/src/asr.rs
+3 -1
View File
@@ -12,7 +12,9 @@
use crate::audio_io::TARGET_SAMPLE_RATE;
use crate::error::{CsmError, Result};
use rubato::{Resampler, SincFixedIn, SincInterpolationParameters, SincInterpolationType, WindowFunction};
use rubato::{
Resampler, SincFixedIn, SincInterpolationParameters, SincInterpolationType, WindowFunction,
};
use std::path::PathBuf;
use std::sync::Mutex;
use whisper_rs::{FullParams, SamplingStrategy, WhisperContext, WhisperContextParameters};